Proceedings - Zurich, Switzerland 20-25 July 2003 ... - IARC Research
Proceedings - Zurich, Switzerland 20-25 July 2003 ... - IARC Research
Proceedings - Zurich, Switzerland 20-25 July 2003 ... - IARC Research
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
8th Infernational Conference on Permafrost<br />
<strong>Zurich</strong>, <strong>Switzerland</strong>,<br />
<strong>20</strong> - <strong>25</strong> <strong>July</strong> <strong>20</strong>03<br />
Extended Abstracts<br />
Reporting Current <strong>Research</strong><br />
and New Information<br />
edited by<br />
Wilfried Haeberli and Dagmar Brandova<br />
Glaciology and Geomorphodynamics Group<br />
Geography Department, University of <strong>Zurich</strong><br />
<strong>Switzerland</strong>
Preface<br />
There is a well-established tradition to prepare and print the <strong>Proceedings</strong> of the International Conferences on Permafrost<br />
(ICOP) in advance, enabling participants to take their volumes home from the meeting. As a consequence of this, papers<br />
have to be submitted, reviewed, corrected and finally edited ahead of the conference and in time for all the work to be<br />
published and delivered. In order to encourage presentation and discussion of ongoing activities and latest results, the<br />
possibility was introduced for the 8th International Conference to present posters on current research and new<br />
information. The present volume contains the extended abstracts of those posters which were personally presented at<br />
ICOP <strong>20</strong>03 in <strong>Zurich</strong>, <strong>Switzerland</strong>, <strong>20</strong> - <strong>25</strong> <strong>July</strong> <strong>20</strong>03. Corresponding 2-page contributions had to be submitted in early<br />
spring <strong>20</strong>03 and were subject to a simplified yeslno-review process by the organizing committee. Our thanks go to the<br />
conference sponsors, to the authors for their excellent collaboration, to Perscheng Assef and Margit Haeberli Vunder for<br />
their help with editing and formatting, to Susan Braun, Marcia Phillips, Sarah Springman and Ivan Woodhatch for their<br />
help with English smoothing, to Andi Kaab for organising the posters and to Martin Holzle for planning and final<br />
preparation of the printing.<br />
Wilfried Haeberli and Dagmar Brandova
Conference Sponsors<br />
Principal Sponsor<br />
International Permafrost Association (IPA)<br />
Co-Sponsors<br />
European Science Foundation (ESF)<br />
International Commission on Snow and Ice (ICSI)<br />
International Glaciological Society (IGS)<br />
International Society for Soil Mechanics and<br />
Geotechnical Engineering (ISSMGE)<br />
Permafrost and Climate in Europe<br />
in the 21st Century (PACE21)<br />
Swiss Academy of Sciences (SANW)<br />
Geoforum <strong>Switzerland</strong> of SANW<br />
Glaciological Commission of SANW<br />
Swiss Geomorphological Society of SANW<br />
Swiss Academy of Engineering Sciences (SATW)<br />
http://www,geodata.soton.ac,uk/ipa/<br />
http://www.esf.org/<br />
http://geowww.uibk.ac.at/research/icsi/<br />
http://www.spri.cam.ac.uk/igs/home.htm<br />
http://w,issmge.org/<br />
http://www.cf,ac.uk/earth/pace/<br />
http://www.naturalsciences.ch/<br />
http://www.geoforum.ethz.ch/<br />
http://www.sanw.glazko.sanwnet.ch<br />
http://www.sanw,ch/exthp/sgmg/<br />
http://w.satw.ch/indexd.html<br />
Funding Sponsors<br />
Main Funding Sponsor<br />
Swiss Agency for Development and Cooperation (DEZA)<br />
Co-funding Sponsors<br />
Airbornescan, <strong>Switzerland</strong><br />
BHP Billiton Diamonds, Canada<br />
Department of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
GeobrugglFatzer Protection Systems, <strong>Switzerland</strong><br />
Helibernina, <strong>Switzerland</strong><br />
Institute for Geotechnical Engineering, ETH <strong>Zurich</strong>, <strong>Switzerland</strong><br />
Migros Culture-Percentage, <strong>Switzerland</strong><br />
Solexperts, <strong>Switzerland</strong><br />
Stump Bohr Drilling Company, <strong>Switzerland</strong><br />
Swiss Academy of Sciences<br />
Swiss Academy of Engineering Sciences<br />
Swiss Alpine Club<br />
Swiss Cablecars<br />
Swiss Federal Institute for Snow and Avalanche <strong>Research</strong><br />
Swiss Federal Office for Water and Geology<br />
Swiss National Cooperative for the Disposal of Radioactive Waste
Extended Abstracts
Kamchatka: mountain permafrost, volcanoes and life<br />
A.A.Abramov, S.N.Buldovich, *V.A.Shcherbakova, K.S.Laurinavichius, **E.M.Rivkina, L.Kholodov and D.A.Gilichinsky<br />
Department of Geocryology, Moscow State University, Moscow, Russia<br />
*Institute of Biochemistry & Physiology of Microorganisms, Russian Academy of Sciences, Pushchino, Moscow Region,<br />
Russia<br />
**Soil Cryology Laboratory, Institute of Physicochemical & Biological Problems in Soil Science, Russian Academy of Sciences,<br />
Pushchino, Moscow Region, Russia<br />
According to the NASA point of view, one way to have liquid<br />
water on Mars at shallow depths would be through subglacial<br />
volcanism. Such volcano-ice interactions are feasible<br />
nowadays beneath the polar ice caps of Mars, or even within<br />
the adjacent permafrost at the margins of the ice caps. This is<br />
why one of the Earth’s models which may come close b<br />
extraterrestrial environments is represented by active volcanoes<br />
in permafrost areas. The main question is whether volcanoes<br />
and associated environments can be ecological niche<br />
for microbial communities, and whether these bacteria can<br />
survive in permafrost. For this reason, our study was carried<br />
out on the volcano Ploskii Tolbachik, situated in Kamchatka -<br />
a unique peninsula in the east of Russia. It is one of the largest<br />
modem volcanic regions with widespread glaciers and<br />
permafrost at high altitudes. Large parts of the glaciers are<br />
well studied, cold and situated at altitudes higher than 3000 m<br />
a.s.1. (Kotlyakov 1997). Permafrost in the region, however,<br />
has received little attention; on the basis of air temperatures,<br />
continuous permafrost can be assumed to occur above 180CL<br />
<strong>20</strong>00 m a.s.1. and to reach a thickness of 50400 m<br />
(Zamolodchikova 1989). In a generalized way, and based on<br />
new data obtained from drilling on the slopes of the volcano in<br />
the summer of <strong>20</strong>02, preliminary results relating to microbiological<br />
and biogeochemical analyses are presented.<br />
One 54 m deep borehole was drilled in 1978 at a site<br />
where the Large Tolbachik Fissure Eruption (LTFE; Fedotov<br />
1984) broke through in 1975176. In addition, we drilled a series<br />
of boreholes and pit-holes at different altitudes between<br />
800 and 2600 m a.s.1.onthe slopes of the active volcano<br />
Ploskii Tolbachik. The boreholes were placed in different<br />
landscape conditions, and had depths ranging from 5 to 15 m.<br />
The permafrost samples for microbiological analysis were<br />
obtained by slow rotary drilling without the use of solutions or<br />
drilling mud. The surface of the extracted cores was trimmed<br />
away with a sterile knife. Then the samples were immediately<br />
plated on a specially-fitted nutrient medium. The main<br />
part of the core was divided into sections, placed in presterilizedaluminumcans,sealedandplacedinfrozenstorage.<br />
Samples were taken for ice content, gas, chemical and<br />
textural composition. After completion of drilling, the boreholes<br />
were closed during a week and then borehole temperatures<br />
were measured.<br />
Based on these field measurements and on other pub<br />
lished data (Zamolodchikova 1989), it is possible to suggest<br />
the following scheme of permafrost distribution on Ploskii Tdbachik<br />
slopes (Fig. 1). The borehole 5/02, located in the forest<br />
zone (-800 m asl.), has no permafrost. We can explain this<br />
by the fact that the trees tend to retain snow. In fact, snow<br />
depths in the forest can reach 2 m and more, whereas in the<br />
absence of trees snow depth averages 0.5 to 0.8 m, and<br />
hardly depends on wind redistribution. Enhanced snow depth<br />
in forests effectively protects the soil surface from strong<br />
winter cooling and enables seasonal freezing only. Borehole<br />
6/02 drilled above the timberline at an altitude of 900 m a.s.1.<br />
is in the zone of seasonal freezing, too. Because of lower<br />
snow cover, however, the remains of seasonally frozen<br />
ground with a thickness of 0.5 m were encountered at depth a<br />
of 1.5 to 2 m.<br />
Borehole “MARS-7/02” was drilled from the surface of a<br />
tefra plateau at about I00 m a.s.1. in the belt of discontinuous<br />
permafrost occurrence. The thickness of the active layer is<br />
about I .5 to 2 m. The mean annual temperature of the frozen
ocks is between 0 and -1" and permafrost thickness up b<br />
50-60 m. Borehole 30 drilled three years after LTFE at about<br />
1000 m a.s.1. revealed 54 m of frozen rocks. It is interesting<br />
to note that the textural description of this borehole indicated<br />
combined piroclastic deposits from LTFE in the upper 16 in m<br />
the summer of 1975. In 1978, these deposits had become<br />
completely frozen (Andreev 1982). Model calculations show<br />
that such deposits with a low heat conductivity need considerable<br />
time for complete freezing (-8-9 yr). We cannot completely<br />
explain this phenomenon yet.<br />
Borehole "PIONER-1/02" was drilled at an elevation d<br />
2600 m a.s.1. from the surface of a weathered basalt flow.<br />
Permafrost here probably has a continuous distribution pattern<br />
and a low mean annual temperature around -7 "C; its thickness<br />
can reach 100-150 m. Local taliks can be found where<br />
thermal waters exist and in the central part of the active volcano.<br />
For better estimates of probable permafrost thickness in<br />
the region, a numerical simulation using the program<br />
"WARM" for the temperature field at the time of LFTE (1975<br />
yr) was conducted. The model dimensions of the Ploskii Tolbachik<br />
volcanic complex are <strong>25</strong> km long and <strong>25</strong> km deep.<br />
The upper boundary condition is given by the mean annual<br />
surface temperature calculated using an altitude lapse rate d<br />
0.5"C/100 m. In glacierized terrain were taken the boundary<br />
condition of Ill type (i.e. combination of mean annual air temperature<br />
and coefficient of heat transfer between atmosphere<br />
and rocks through ice with an estimated thickness of 100 m).<br />
At the lower boundary, located on the depth <strong>25</strong>000 m, was<br />
set heat flux 80 mW/m2. On the depth <strong>20</strong>000 m was set the<br />
magma reservoir (3 km in diameter), with temperature<br />
1<strong>20</strong>0°C. Thermal properties of the rocks were taken from the<br />
literature data (Komarov et al. 1996; Table I).<br />
Table 1. Thermal properties of rocks used for simulation.<br />
Basalt Volcanic dross<br />
AT, W/m* K 2.9 0.3<br />
hf. W/m*K 3.5 0.6<br />
Cr, Kkal/m'*K 600 300<br />
Cp, Kkal/m3*K 500 400<br />
Q~, ua1lrn3 5000 1000<br />
With the results of the simulation it is possible to conclude<br />
that the volcanic eruption causes instantaneous temperature<br />
rises of up to 1000-1<strong>20</strong>0") but constitutes short-lived effects.<br />
The main effects are concentrated at the surface and within<br />
high atmospheric layers. Thus, it is not likely that such erup<br />
tions will greatly reduce the thickness of the permafrost, any<br />
effects will remain localized. Increased heat flux in areas d<br />
modem volcanic activity, on the other hand, results in considerable<br />
reduction of permafrost thickness underneath glaciers,<br />
thereby probably inducing the formation of zones with<br />
positive subglacial temperatures and basal glacier melting.<br />
The subsurface temperature field at the the Ploskii Tolbachik<br />
volcano has a non-stationary character, because it could not<br />
fully adjust between eruptions.<br />
Abramov, A.A. et al.: Kamchatka: mountain permafrost, volcanoes and ...<br />
The existence of frozen rocks on the investigated volcano<br />
is a function of elevation rather than latitude. As a consequence,<br />
it should be attributed to mountain permafrost conditions.<br />
In the investigated region, continuous permafrost with a<br />
thickness of about 50 m starts to occur at an elevation higher<br />
than 1400-1500 m a.s.1. Maximum permafrost thickness can<br />
probably reach <strong>20</strong>0-<strong>25</strong>0 m at 3000 a.s.1. m<br />
The analysis shows thatfrozensamplesextractedfrom<br />
borehole 7/02 and representing young volcanic deposits<br />
contain viable microorganisms and, among them, therm@<br />
philic anaerobic bacteria. Moreover, biogenic methane (up b<br />
1100-1900 pICH4lkg soil) was found in these samples.<br />
Thermophiles had never been found before in permafrost. The<br />
present study therefore demonstrates that the only way for<br />
thermophilic bacteria to appear within frozen volcanic horizon<br />
is during eruption from volcanoes or from related subsurface<br />
geological strata. The most important conclusion is that (1)<br />
thermophilicbacteriamight survive in permafrost and even<br />
produce biogenic gases, and that (2) terrestrial volcanic microbial<br />
communities might serve as exobiological models for<br />
hypothesesonexistingancientmicrobiocenoses,i.e., extraterrestrial<br />
habitats that may possibly be found under anoxic<br />
conditions around volcanoes on Mars or other planets.<br />
ACKNOWLEDGEMENTS<br />
This work was supported by the Russian Foundation of Basic<br />
<strong>Research</strong> (grant 01-05-05-65043) and the NASA Astrobiology<br />
Institute Program.<br />
REFERENCES<br />
Andreev, V. 1982. Permafrost in Tolbachik eruption region. Questions<br />
of Kamchatka geography 8: 98-99 (in Russian).<br />
Fedotov, S. (ed.) 1984. Large Tolbachik Fissure Eruption Kamchatka<br />
1975-1976. Moscow: Nauka (in Russian).<br />
Komarov, I. et al 1996. Heat- and mass-exchange properties of<br />
rocks. In: E. Ershov (ed.), Lithogenous Geocryology. Moscow<br />
State University publisher (in Russian).<br />
Kotlyakov, V. (ed.) 1997. World atlas of snow and ice resources .<br />
Moscow: Nauka (in Russian).<br />
Zamolodchikova, S. 1989. Kamchatka region. In: Ershov, E. (ed.),<br />
Geocryology of USSR. East Siberian and far East: 380-389.<br />
Moscow: Nedra (in Russian).<br />
2
8th international Conference on Permafrost Extended Abstracts an Current <strong>Research</strong> and Newly Available Information<br />
Verifying modelling approaches: high mountain permafrost and its<br />
environment<br />
M. Avian<br />
Department of Geography and Regional Sciences, University of Graz, Austria<br />
INTRODUCTION<br />
The Ankogel Mountains are a highly glacierized area in the<br />
easternmost part of the Hohe Tauern Range in Carinthia<br />
(Fig. I). The test site encloses 98 km* including the main<br />
peaks of the Ankogel (3246 m) and the Hochalmspitze<br />
(3360 m). Glaciers cover an area of 12.1 km2 and shows<br />
wide retreat zones.<br />
Remote sensing data are the basis for the mapping of<br />
the distribution of snow patches, vegetation and rock glaciers.<br />
Snow patches were mapped by using colour orthophotographs<br />
(as at: August 1998), Keller (1994) shows an<br />
investigation, in which a significant correlation between<br />
493 snow patches and the occurrence of permafrost was<br />
detected. Infrared orthophotographs are a common data<br />
set for investigating vegetation. A useful classification in<br />
connection with the investigation of permafrost could only<br />
be carried out by visual interpretation. In connection with<br />
their percentage of soil cover, three classes of vegetation<br />
were used: uniform vegetation (over go%,), transition<br />
vegetation (<strong>20</strong> to go%,), and island vegetation (below<br />
<strong>20</strong>%). The background is the assumption that a solid cover<br />
of vegetation excludes permafrost in the underground. The<br />
occurrence of active rock glaciers is an obvious sign for<br />
permafrost.<br />
RESULTS<br />
Figure 1. Location of the study area in Austria.<br />
METHODOLOGY<br />
The distribution of permafrost was modelled by an adaptation<br />
of the programme PERMAKART, using the GIS Arc-<br />
Info (Haeberli et al. 1996). The main point is to receive information<br />
about slope, aspect and elevation. The result is a<br />
supposed distribution of permafrost in the study area with<br />
the classes “permafrost probable”, “permafrost possible”<br />
and “no permafrost”. The modelling of potential incoming<br />
short wave radiation (PSWR) was done by using the programme<br />
ERDAS-Imagine. Shadows that are cast by topographic<br />
features on the surrounding surface were not calculated.<br />
As the distribution of exposition is homogenous it is a good<br />
basis for further analysis. The modelled classes of the distribution<br />
of permafrost were renamed for the reason of empirical<br />
results in the Hohe Tauern Range (Lieb 1998) from<br />
“permafrost possible” to “potential sporadic permafrost<br />
(PSP)” and from “permafrost probable” to “potential discontinous<br />
permafrost (PDP)” (see Tab. 1). The model of<br />
the PSWR shows a satisfying overall view and a good correspondence<br />
of areas with a PSWR under 53 kJ/m2a and<br />
PDP.<br />
The main part of the study is to compare the results of<br />
the modelling with the information received from the study<br />
area. 684 snow patches with an area of altogether 108 ha<br />
were counted in the study area (average altitude: 2635m).<br />
At first sight there is no significant relation between PDP<br />
and the occurrence of snow patches (Tab. 2), but a closer<br />
look shows that nearly all snow patches which are not situated<br />
in potential permafrost areas, occur in direct neighborhood<br />
with PDP and PSP.<br />
3
29.3% of the study area are covered with vegetation,<br />
the relation between potential permafrost and vegetation<br />
provides a clear result: although there is a high percentage<br />
of soil cover with vegetation in this high mountain area,<br />
only a marginal zone of overlapping with PDP in an area of<br />
0.9 hectares and 0.06% can be seen (Table 3). Rock glaciers<br />
can only be found in the southern part of the study<br />
area. Seven are active, one is inactive and eight are fossile.<br />
The model shows a good correlation with the catchment<br />
area of blocks, but does not mainly correspond with<br />
the lobes of the rock glaciers.<br />
Table 1. Areas of generated permafrost classes.<br />
area in km2 area in %<br />
no permafrost<br />
PDP 24.0 24.6<br />
sum 97.6 100.0<br />
Table 2. Snow patches according to potential permafrost areas.<br />
Avian, M.: Verifying m ~ approaches: ~ high ~ mountain ~ permafrost.. l ~ ~ ~<br />
1 area in km2 I area in %<br />
no nermafrnst I 0.37 "" I 84.5 - ."<br />
PSD 0.37 34.0<br />
PDP 0.34 31.5<br />
sum I 1.08 I 100.0<br />
Table 3. Areas of vegetation classes and overlapping with potential<br />
permafrost areas.<br />
area in area in Overlapping %<br />
km* % PDP PSP<br />
uniform veg. 17.9 62.3 0.05 1.33<br />
transition veg 8.6 0.12 30.1 9.86<br />
island veg. 2.1 2.30 7.6 19.06<br />
sum 28.6 100.0<br />
The final step is to generate a most probable lower borderline<br />
for permafrost including all results with a last visual<br />
interpretation of the geological situation (coarse debris).<br />
The adaptation uses the modelling of potential permafrost<br />
areas as a basis, the results of the distribution of vegetation<br />
correct the limit as well as the distribution of snow<br />
patches. A visual interpretation of the orthophotographs includes<br />
areas of coarse debris in the neighborhood of potential<br />
permafrost areas. The modelling of PSWR is only a<br />
check in a higher scale. The final result is an area of 35.16<br />
km2, that are 36% of the study area, which can be seen as<br />
potential permafrost areas. In April <strong>20</strong>02 BTSmeasurements<br />
were carried out in two small test sites in<br />
the south east of the study area. The results correspond<br />
very well with the most likely lower borderline for permafrost.<br />
I. Although there is a superficial discrepancy between<br />
the assumption and mapping results of<br />
snow patches, a closer look provides valuable improvement.<br />
2. Closed vegetation cover excludes potential permafrost<br />
in the study area with a probability of<br />
99,96%.<br />
3. All catchment areas of active rock glaciers are<br />
situated in potential permafrost areas.<br />
4. The result of modelling the potential incoming<br />
short wave radiation shows a good connection to<br />
potential permafrost.<br />
Crosschecking of results shows good correlations between<br />
the modelling results and the results from mapping<br />
for the Eastern part of the Alps. Valuable improvements<br />
can be derived in using the distribution of snow patches<br />
and geological patterns.<br />
ACKNOWLEDGEMENTS<br />
I wish to thank Mr. Gerhard Lieb and Ms. Michaela Nutz for<br />
critical review, also Mrs. Christine Lazar for improving the<br />
English and Mr. Stefan Lieb, Mr. Markus Frei, Mr. Wolfgang<br />
Sulzer as well as Mr. Alexander Avian for critical<br />
comments.<br />
REFERENCES<br />
Haeberli, W., Hoelzle M., Dousse, J.-P., Ehrler, C., Gardaz, J.-M., Imhof,<br />
M., Keller, F., Kunz, p., Lugon, R. and Reynard, M. 1996.<br />
Simulation der Permafrostverbreitung in den Alpen mit geographischen<br />
lnformationssystemen. Arbeitsbericht NFP 31,<br />
Hochschulverlags-AG an der ETH <strong>Zurich</strong> ,58 p.<br />
Keller, F. 1992. Automated mapping of mountain permafrost using the<br />
programm PERMAKART within the Geographical Information<br />
System ARC-Info. Permafrost and Periglacial Processes, Vol. 3(2):<br />
133-138.<br />
Keller, F. 1994. lnteraktionen zwischen Schnee und Permafrost, Eine<br />
Grundlagenstudie im Oberengadin. Versuchsanstalt fur Wasserbau,<br />
Hydrologie und Glaziologie der ETH <strong>Zurich</strong>, 145 p,<br />
Lieb, G.K. 1998. High-mountain permafrost in the Austrian Alps. -<br />
Permafrost. 7th International Conference on Permafrost. Yellow-<br />
CONCLUSION<br />
The model of the distribution of permafrost provides wide<br />
areas of potential permafrost, including almost the entire<br />
glacier areas. All the relations can be summarised as follows:
The role of cosmoplanetary climate cycles in Earth’s cryolithosphere<br />
evolution<br />
V. T. Balobaev and V. V. Shepelev<br />
Melnjkov Permafrost Institute, $0 RAS, Yakutsk, Russia<br />
We believe that different climate-forming global factors<br />
have a resonance type of relationship. The conceptual<br />
model of manifestation of this effect on the global climate<br />
system is shown in Figure 1. This model is based on the<br />
assumption that periods and amplitudes of the temperature<br />
variations caused by separate global factors are constant<br />
in time. The presented model takes into account four<br />
main climate-forming factors: one cosmoplanetary variation<br />
with a period of <strong>20</strong>0 Ma and three astroplanetary<br />
variation with periods of 100, 40.7, and <strong>20</strong> ka. The variation<br />
curve of the longer cycle is taken as the base level for<br />
variations of the shorter climate-forming cycle. The superposition<br />
of harmonics involved in the calculation of<br />
climate-forming cycles highlights distinct peaks corresponding<br />
to periods of phase coincidence of individual temperature<br />
variations. This effect, which we called thermoresonance,<br />
causes significant periodic cooling and warming<br />
of the global climate.<br />
(k<br />
k.<br />
E 4<br />
lntcrval helwecn extrcrnc wanning pcnuds. ka<br />
40 **<br />
80 ,<br />
I<br />
Figure 1. Conceptual model of the integral dynamic correlation<br />
between the main global factors of climate formation during the past<br />
150 ka. (1) curve of a <strong>20</strong>-ka cycle caused by precession of the Earth’s<br />
axis; (2) curve of a 40.7-ka cycle caused by variations in the obliquity<br />
of the rotation axis with respect to the ecliptic plane; (3) curve of a<br />
100-ka cycle caused by variations in the eccentricity of the Earth’s<br />
orbit: (4) curve of an about <strong>20</strong>0-Ma cycle related to the rotation of the<br />
solar system around the galactic centre; (5) area of optimal climatic<br />
conditions; (6) periods of extreme warming; (7) periods of extreme<br />
cooling.<br />
5<br />
The amplitudes of harmonics strongly vary and the<br />
highest of them, especially those of 40-ka and <strong>20</strong>-ka<br />
cycles, distinctly indicate the resonance amplification.<br />
Since the Earth’s climate system is excited and less stable<br />
during the strongest temperature decrease or increase in<br />
the 100-ka cycle, the 40- and <strong>20</strong>-ka cycles exert a<br />
resonance effect. It should be noted that extreme cold<br />
periods in our model generally agree with glacial periods<br />
in the Late Pleistocene reconstructed from seafloor<br />
sediment data (Steig 1998).<br />
Using the proposed concept of the thermoresonance<br />
effect, an attempt can be made to reconstruct the climate<br />
for the whole Phanerozoic Eon spanning the past 570 Ma.<br />
Figure 2 shows a scheme constructed with consideration<br />
for the thermoresonance effect between two main<br />
cosmoplanetary climate-forming factors with periods of 1.3<br />
Ga and <strong>20</strong>0 Ma and spanning the past 700 Ma of the<br />
Earth’s evolution.<br />
Of special interest for permafrost-researchers is to<br />
reveal the possibility of formation and dynamics of<br />
cryogenic processes in different periods of the<br />
Phanerozoic which was relatively warm in the Earth’s<br />
history that had served as the base for a vigorous<br />
development of a biosphere. The preceded Proterozoic<br />
was significantly colder, what determined the existence of<br />
large glaciations whose traces are found in sediments of<br />
this epoch. The Cambrian (570-500 Ma ago), that is the<br />
first period of the Phanerozoic, was warm and without<br />
permafrost what is connected not only with <strong>20</strong>0-Ma cycle<br />
(Fig. 2) but with favourable location of lithospheric plates<br />
on the Earth’s surface as well. The largest continent of<br />
that time, Gondwana, was located in subtropical latitudes,<br />
and smaller plates were in the equator zone. The poles<br />
were located in water areas of open oceans, and negative<br />
temperatures were not observed on the Earth’s surface of<br />
the whole planet.<br />
In the Ordovician (500-440 Ma ago) the planetary<br />
climate was significantly colder (Fig. 2). By the end of this<br />
period, Gondwana transferred to the South Pole. This<br />
determined the initiation of a thick continental glaciation,
Balobajev, V.T. et al.: The role of cosrnoplanetary climate G~GIES..<br />
whose traces occur in Africa. Frozen grounds developed<br />
at a third part of Gondwana (Africa and South America).<br />
Figure 2. Scheme of the correlation between two main cosmoplanetary<br />
factors responsible for the Earth's global climate. (1) curve<br />
of the 1.3-Ga cycle caused by rotation of our galaxy around the megagalactic<br />
centre: (2) curve of a <strong>20</strong>0-Ma cycle caused by rotation of the<br />
solar system around the galactic centre; (3) area of optimal climatic<br />
conditions: (4) extreme warming period; (5) extreme cooling periods.<br />
The water areas of the North Pole could have seasonal<br />
icings only. The global warming occurred in the Silurian<br />
(440-410 Ma ago). The Gondwana continent left the South<br />
Pole and transferred to the equator. Thin frozen grounds<br />
could occur just on a small part of this continent (South<br />
America). The area of their development and the thickness<br />
increased by the end of the Silurian due to drifting of<br />
Gondwana to the South Pole.<br />
The cycle of global warming (440-300 Ma ago)<br />
coinciding with the periods of the Devonian and the<br />
Carboniferous was one of the warmest cycles on the<br />
Earth. It was accompanied by a vigorous development of<br />
vegetation. The addition of small lithospheric plates to the<br />
second supercontinent, Laurasia, occurred in this epoch,<br />
and at the end of this epoch Gondwana joined with<br />
Laurasia into a single supercontinent Pangea. There was<br />
no glaciation though, Gondwana and later a significant<br />
part of Pangea were located near the South Pole<br />
throughout this period. However, considerable daily and<br />
annual fluctuations of temperature and its transitions<br />
through 0°C was to be observed due to a large area of the<br />
continent and location of its parts in high latitudes. At the<br />
end of the Carboniferous (350-285 Ma ago), when the<br />
global temperature began to decrease, negative mean<br />
annual temperatures and frozen grounds appeared in the<br />
south part of Pangea. Then the continental glaciation<br />
began near the South Pole (Antarctica, South Africa,<br />
Australia). The strongest cooling of the Earth<br />
wasobsewed at the boundary of the Permian (285-230 Ma<br />
ago) and the Triassic (230-195 Ma ago). It was connected<br />
with the global climate cooling and unfavourable location<br />
of Pangea relative to the poles. The south of Pangea was<br />
covered with ice (Antarctica), and the south of the<br />
continent contained permafrost.<br />
The subsequent cycle of a global warming occurred in<br />
the Jurassic (195-137 Ma ago) and the Cretaceous (137-<br />
67 Ma ago, Fig. 2). By that time Pangea was divided into<br />
Gondwana and Laurasia and then into smaller blocks, of<br />
which present continents were formed later. The reasons<br />
that Gondwana moved to the latitudes of <strong>20</strong>" from the<br />
South Pole and that both poles of the Earth occurred in<br />
the ocean resulted in a greater increase of the global<br />
temperature. Mean global temperature at that time was<br />
8°C higher than the present. All of this eliminated the<br />
possibility of formation of permafrost on the Earth and<br />
glaciation centres.<br />
At the end of the Cretaceous, a new cold epoch developed<br />
on the Earth. Compared to the Late Cretaceous, temperature<br />
decreased by 2-4°C at the middle of the Paleogene<br />
(65-23 ka) and was 44°C higher than the present at<br />
the end of the Paleogene. Fast decrease of the global<br />
temperature was due to the fact that Antarctica firmly<br />
established at the South Pole. Formation of the Antarctic<br />
circumpolar stream at the beginning of the Neogene<br />
resulted in glaciation of Antarctica, which continues to the<br />
present (Monin and Shishkov 1979). In the northern<br />
hemisphere first sheet glaciations are dating by 3 Ma ago,<br />
and the ice cover in the Arctic Ocean appeared 0.7-0.8 Ma<br />
ago. Probably, at the same time the permafrost formed in<br />
the northern hemisphere.<br />
Therefore, positive phase of the 1.3-Ga climate cycle,<br />
including practically the entire Phanerozoic, is terminating.<br />
This indicates that the next 500-600 Ma on our planet will<br />
be a kingdom of cold and ice. However, this does not<br />
exclude short-term warming periods related to astro- and<br />
geoplanetary temperature-forming factors and their resonance<br />
amplification.<br />
REFERENCES<br />
Monin, A.S. and Shishkov, Yu.A. 1979. Climate history (Istoriya<br />
klimata) (in Russian). Leningrad: Gidrometeoizdat.<br />
Steig, E. 1998. Synchronous climate changes in Antarctica and the<br />
North Atlantic. In Science 282: 92-95.<br />
6
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Terrestrial biosphere modelling in permafrost regions<br />
C. Beer, *W. Lucht, **C. Schmullius and M. Gude<br />
Department of Geography, University of Jena, Germany<br />
*fofsdam Institute for Climate lmpact <strong>Research</strong>, Potsdam, Germany<br />
**Department of Geography, University of Jena, Germany<br />
INTRODUCTION<br />
Besides C02 inversion modelling, Dynamic Global Vegetation<br />
Models (DGVMs) are suited for investigating the terrestrial<br />
component of the global carbon balance. These<br />
models are icreasingly of particular interest due to their<br />
potential for predicting the impacts of the biosphere upon<br />
future concentrations of atmospheric C02. Because water<br />
availability in plants is one of the key stimuli for vegetation<br />
growth, DGVMs combine the carbon with the water cycle.<br />
Our research is conducted within the framework of the<br />
SIBERIA-II project funded by the European Commission<br />
(Contract No. EVG2-<strong>20</strong>01-00008, http://www.siberia2.unijena.de).<br />
In this work, the impact of hydrological conditions<br />
on the carbon cycle in permafrost regions of Northern<br />
Eurasia is investigated. For this purpose, a permafrost<br />
module built upon previous work of Venevsky (<strong>20</strong>01) following<br />
the equations of Anisimov et al. (1997) is presented<br />
as an extension of the Dynamic Global Vegetation Model<br />
LPJ (Lund-Potsdam-Jena; Sitch <strong>20</strong>00).<br />
MODEL DESCRIPTION<br />
LPJ uses 4 carbon and 2 soil layers (0.5 m and IS m<br />
depth). Input of LPJ are basic climate and soil data (air<br />
temperature, precipitation, sunshine duration, soil texture)<br />
on a spatial resolution of 0.5" times 0.5". To provide dynamic<br />
variables of the carbon cycle such as Net Primary<br />
Production (NPP) or Leaf Area Index (LAI) in a daily time<br />
step, LPJ begins by computing hydrological variables such<br />
as soil moisture and evapotranspiration.<br />
In permafrost regions the hydrology is completely altered<br />
by our new permafrost module. Soil moisture is<br />
strongly dependent on daily thaw depth in the present. As<br />
a simplification, we assume that the course of daily thaw<br />
depth and air temperature resembles a sine curve. This<br />
leads to the following equation for calculating the daily<br />
thaw depth:<br />
Zthaw - daily thaw depth<br />
L a x - max thaw depth<br />
t - current day<br />
tspring - start of spring<br />
zpos - duration of days while T > 0<br />
T - air temperature<br />
The zonation of permafrost (Fig. 1) is obtained by assuming<br />
that TTOP (Temperature at TOP Of the Permafrost)<br />
is below 0 "C in permafrost regions following Smith<br />
(<strong>20</strong>02). This zonation will be compared with that obtained<br />
by means of the frost-index of Anisimov and Nelson<br />
(1 997).<br />
OUTLOOK<br />
To investigate the impact of permafrost hydrology on the<br />
terrestrial carbon cycle, different outputs of LPJ with and<br />
without the permafrost module are compared with in situ<br />
measurements or Earth Observation data, e. g., growing<br />
stock volume, NPP, or LA1 in Siberia between 90" and<br />
I IO"East. A lower value of biomass and NPP is expected<br />
when permafrost hydrology is taken into consideration,<br />
and it is anticipated that more reliable values for NPP and<br />
NEP (Net Ecosystem Production), can be obtained, which<br />
in turn will lead to a more precise global carbon balance.<br />
7
Beer, 6. et al,: Terrestrial biosphere modelling in ~ ~r~~~fr~st<br />
regions..<br />
REFERENCES<br />
Anisimov, O.A. Shiklomanov, N.I., and Nelson, F.E. 1997. Global<br />
warming and active-layer thickness: results from transient general<br />
circulation models. Global and Planetary Change 15: 61-77.<br />
Anisimov, O.A. and Nelson, F.E. 1997. Permafrost zonation and climate<br />
change in the northern hemisphere: results from the transient<br />
general circulation models, Climatic Change 35: 241-<strong>25</strong>8.<br />
Sitch, S. <strong>20</strong>00. The role of vegetation dynamics in the control of atmospheric<br />
COe content. Dissertation, Lund University, Lund,<br />
Sweden<br />
Smith, M. W. and Riseborough, D.W. <strong>20</strong>02. Climate and the limits of<br />
permafrost: A Zonal Analysis. Permafrost and periglacial Processes<br />
13: 1-15.<br />
Venevsky, S. <strong>20</strong>01. Broad-scale vegetation dynamics in north-eastern<br />
Eurasia - observations and simulations, PhD Thesis, Universitat<br />
fur Bodenkultur. Wien.
Combining water chemistry and geophysical investigations with assessment<br />
of chemical denudation rates in the Latnjavagge drainage basin, arctic-oceanic<br />
Swedish Lapland<br />
A. A. Beylich, E. Kolstrup, *U. Molau, **T. Thyrsfed, * W. Linde, L. B. Pedersen and ****D. Gintz<br />
Department of Earth Sciences, Environment and landscape Dynamics, Uppsala University, Uppsala, Sweden<br />
*Botanical Institute, Plant Ecology, Gothenburg University, Gothenburg, Sweden<br />
**Harbacken, Stavby, Alunda, Sweden<br />
***Department of Earth Sciences, Geophysics, Uppsala University, Uppsala, Sweden<br />
““*Institute for Geological Sciences, AB Hydrogeology, Free University of Berlin, Berlin, Germany<br />
In I999 a monitoring programme was started in an arcticoceanic<br />
drainage basin, Latnjavagge (9 km2, 950 - 1440<br />
m a.s.1.; 68”<strong>20</strong>’N, 18”30‘E), with the objective to quantify<br />
the sediment budget and the relevant denudation processes<br />
as dependant on environmental parameters<br />
(Beylich in press). With regard to chemical denudation<br />
rates several factors such as topography, climate, lithology,<br />
regolith thickness and ground frost are of importance,<br />
so that as a consequence, different methods need to be<br />
combined in order to assess the chemical denudation.<br />
Once an integrated approach is applied, it is possible to<br />
analyze chemical denudation rates for clearly-defined<br />
landscape units such as confined fluvial drainage basins<br />
(Beylich <strong>20</strong>02).<br />
The Latnjavagge drainage basin in the periglacial<br />
Abisko mountain area is characterized by a homogeneous<br />
lithology (mica shists). Water balance, water chemistry<br />
and radio magnetotelluric geophysical investigations were<br />
integrated with assessment of regolith thickness along<br />
selected profiles as well as of ground frost conditions<br />
within the catchment. In combination with direct field observations,<br />
the geophysical profiles demonstrated the<br />
presence of very thin regolith in most of the investigated<br />
areas, yet in some parts the bedrock was located deeper<br />
and was not detected locally at 40 m depth. The low resistivities<br />
found along the profiles in the geophysical investigation<br />
showed that frozen ground was not found around<br />
1000 m a.s.1. in late August (Beylich et a/. subm.b). Permafrost<br />
in the area seems to be restricted to areas above<br />
1000 m a.s.1. Water chemistry and total dissolved solids<br />
(TDS) of precipitation, snowpack and surface water have<br />
been analyzed on samples collected at <strong>25</strong> selected sites<br />
within Latnjavagge (Beylich et a/. subm-a). Additionally,<br />
daily discharge and yield of dissolved solids have been<br />
calculated for several subcatchments (Beylich et a/.<br />
submx). Water laboratory analyses of <strong>20</strong>5 samples’taken<br />
in the field included the components Ca2*, Mgzt, Na+, Kt,<br />
Fe2+, Mn2+, CI-, Nos-, so42- and Po$, with S042- and Can+<br />
being the dominant ones in the surface water (Beylich et<br />
a/. subm.a). The salt concentrations and chemical denudation<br />
in this arctic-oceanic periglacial environment were<br />
found to be low with a mean annual chemical denudation<br />
rate for the entire catchment of 5.4 ff(km2 yr) (Beylich et a/.<br />
subm.c).<br />
Comparison of the water chemistry data from different<br />
sampling sites and subcatchments by means of cluster<br />
analysis shows that there are significant differences within<br />
the area: In areas of shade, longer snow cover, frozen<br />
ground and thin regolith, the salt concentrations over the<br />
summer were perceptible but so low that salts brought into<br />
the basin from precipitation and melting snow could be<br />
clearly detected in the surface water. The highest salt<br />
concentrations were found at a radiation-exposed west<br />
facing, vegetated, moderately steep slope with relatively<br />
thick regolith that was thawed already from the time of<br />
snowmelt in early June. In such well-drained sites with<br />
continuous subsurface water flow a maximum of contact<br />
between water and mineral particles can take place<br />
(Beylich et a/. subm.a). The concentration values revealed<br />
differences in the rate of thawing of frozen ground between<br />
shaded areas andlor areas at higher altitude on the<br />
one hand, and radiation-exposed areas on the other. A<br />
comparison with the results from Karkevagge a few kilometres<br />
to the northwest as well as from other periglacial<br />
locations indicates that the chemical denudation values<br />
from Latnjavagge are fairly representative of periglacial<br />
oceanic environments, more so than the values from the<br />
Karkevagge catchment (Beylich et a/. subm.a). The investigation<br />
in Latnjavagge stresses the importance of spatial<br />
variability within even small catchments of homogeneous<br />
lithology as it demonstrates that salt concentrations from<br />
different subareas can differ substantially depending on<br />
exposure to radiation, duration of snow cover and frozen<br />
ground conditions, regolith thickness, and also on vegetation<br />
cover and slope angle as factors steering water turbulence<br />
and retention of drainage (Beylich et a/. subm-a).<br />
9<br />
REFERENCES<br />
Beylich, A. A,, <strong>20</strong>01. Slope denudation, streamwork and relief development<br />
in two periglacial environments in East Iceland and<br />
Swedish Lapland. Transactions, Japanese Geomorphological<br />
Union, 22 (4), C-23
Beylich, A. A. et al.: Combining water chemistry and ~ ~ ~ investigations ~ ~ ~ s...<br />
i c ~ l<br />
Beylich, A.A. in press. Sediment budgets and relief development in<br />
present periglacial environments - a morphosystem analytical approach,<br />
Hallesches Jahrbuch fur Geowissenschaften, A.<br />
Beylich, A.A., Kolstrup, E., Thyrsted, T., Gintz, D., subm.a. Water<br />
chemistry and its diversity in relation to local factors in the<br />
Latnjavagge drainage basin, arctic oceanic Swedish Lapland,<br />
Geomorphology.<br />
Beylich, A.A., Kolstrup, E., Thyrsted, T., Linde, N., Pedersen, L.B.,<br />
Dynesius, L., subrn-b. Chemical denudation in arctic-alpine<br />
Latnjavagge (Swedish Lapland) in relation to regolith thickness as<br />
assessed by radio magnetotelluric (RMT)-geophysical profiles,<br />
Geomorphology.<br />
Beylich, A.A., Molau, U., Luthbom, K., Gintz, D., subm.c. Rates of<br />
chemical and mechanical fluvial denudation in an arctic-oceanic<br />
periglacial environment, Latnjavagge drainage basin, northern-
8th International Conference on Permafrost Extended Abstracts on Current. <strong>Research</strong> and Newly Available i n f ~ ~ ~ ~ ~<br />
Analysis of permafrost thickness in Antarctica using VES and MTS data<br />
E. Borzotta and *D. Trornbotto<br />
Unidad de Geofisica, *Unidad de Geocriologia<br />
lnstitufo Argentino de Nivologia, Glaciologia y Ciencias Ambientales (IANIGLA)<br />
CONICET, Casilla de Correo 330, 5500 Mendoza, Argentina<br />
Frozen ground thicknesses measured in the NE of the<br />
Antarctic Peninsula (Seymour and James Ross Islands;<br />
Fig. I) and estimations of permafrost thicknesses for the<br />
same areas are being compared and analyzed using (a)<br />
vertical electrical soundings data (VES), (b) mean annual<br />
air temperatureland (c) steady geothermal heat flows as<br />
inferred from magnetotelluric soundings (MTS).<br />
In the present study, as part of a more exhaustive<br />
study in progress, estimations of steady heat flows are<br />
made, which could contribute to a better understanding of<br />
the cryogenic processes in the area.<br />
Figure 1. Antarctic Peninsula. The frames indicate the zones where<br />
permafrost studies were performed (Seymour and James Ross Islands).<br />
Seymour and James Ross Islands are located in the<br />
Weddell Sea, next to the Antarctic Peninsula, at about<br />
64"s and 57"W. Both are part of the Larsen Basin located<br />
eastward of the Antarctic Peninsula. Upper Cretaceous<br />
marine sediments outcrop in the south portion of Seymour<br />
Island and Tertiary sediments at the north. This island is<br />
free of ice cover. James Ross Island is a Plio-Pleistocene<br />
volcanic island with recent activity at its eastern border<br />
and around 70% of its surface is covered by glaciers. At<br />
Seymour Island, a mean annual air temperature (MAAT)<br />
of -9.4"C was estimated at <strong>20</strong>0 m a.s.1. (upper terrace)<br />
from a ten-year temperature recording (1971-1980). Temperatures<br />
at 10 m depth were measured during the 1976-<br />
1981 period in the snow on James Ross Island (Dalinger<br />
rl"S<br />
Dome and Mt. Haddington), which let to estimate a MAAT<br />
of -13°C at about 1410 m, on average, and -5°C at 35 m<br />
a.s.1. where geophysical studies were conducted later by<br />
Fukuda and others (Fukuda et al. 1992).<br />
-The first magnetotelluric soundings (MTS) in Seymour<br />
Island were carried out in its northern section (upper terrace)<br />
in 1979 and 1980 (Fournier et al. 1980, Del Valle et<br />
al. 1982), describing the Larsen Basin only below 600 m<br />
depth; thus it was not possible to study the permafrost.<br />
Results showed a very conductive layer (saline) below the<br />
frozen ground with 40 Siemens of conductance. A basin<br />
basement at 5.2 km depth is also shown by these data<br />
and a conductive layer (possible asthenosphere) is suggested<br />
in the upper mantle at longer periods with a top at<br />
63 km depth.<br />
In 1992, three MT soundings were made in James<br />
Ross Island (Brandy Bay and Hidden Lake). Results<br />
showed a conductance below the frozen ground one order<br />
of magnitude higher than below the frozen ground at<br />
Seymour Island: 40 Siemens in Seymour Island, 315 Siemens<br />
in James Ross Island. As in the case of Seymour<br />
Island, a possible asthenosphere is also suggested for<br />
James Ross Island with a top at 61 km depth. Distortions<br />
present in the apparent resistivity curves and the volcanic<br />
nature of this island, with recent activity, lead us to suspect<br />
the existence of a possible magma chamber seated<br />
between 6.8 km and 14 km depth in the crust at NW of<br />
James Ross Island. These electromagnetic studies performed<br />
in Seymour and James Ross Islands were not able<br />
to describe the permafrost directly.<br />
Vertical electrical soundings (VES), particularly dedicated<br />
to studying the permafrost, were conducted on<br />
Seymour and James Ross Islands in the summer 1989-<br />
1990 (Fukuda et al. 1992). These studies were carried out<br />
at different terraces observed by these authors. In<br />
Seymour Island, <strong>20</strong>0 m of frozen ground thickness was<br />
estimated in the upper terrace (around <strong>20</strong>0 m a.s.l.), 105m<br />
in the middle terrace (around 50 m a.s.1.) and 35m in lower<br />
terrace at 5 m a.s.1. Maximum resistivities of about 7<strong>20</strong> to<br />
800 8m were measured in frozen ground by these<br />
authors (Fukuda et al. 1992). In a general context, the<br />
thickness of the active layer in Seymour Island could be<br />
11
estimated at 0.35 to 0.60m. In the same campaign, four<br />
VES were carried out in the ice-free northwestern region<br />
of James Ross Island. Shallower thicknesses of frozen<br />
ground than those estimated at Seymour Island were estimated<br />
in this case: 40 to 45 m in the upper terraces<br />
(around 21 to 35 m a.s.l.), 3.4 m in the middle terrace<br />
(around IOto 17 m a.s.1.) and 5.8 m in the lower terrace at<br />
about 3 to 5 m a.s.1. Resistivities between 1000 and 4000<br />
m were obtained, with a thickness of the active layer<br />
approximately 0.70 to 1.45 m (Fukuda et al. 1992).<br />
From the depths of conductive layers in the crust and<br />
the upper mantle determined by MT soundings, a gross<br />
estimation of the regional geothermal heat flow - in steady<br />
condition - may be obtained, and consequently, the geothermal<br />
gradient and the permafrost thickness in equilibrium<br />
to the present MAAT can be estimated. In the present<br />
current study at Seymour and James Ross Islands,<br />
the frozen ground thickness - measured using VES - and<br />
the estimated permafrost thickness, obtained considering<br />
homogeneous medium and steady heat flows inferred<br />
from MT soundings results, are being compared and analyzed.<br />
Empirical exponential formulas correlate the geothermal<br />
heat flows and the depths of conductive layers determined<br />
using MTS (Adam 1978). These expressions are<br />
based principally on information published in Adam<br />
(1976), corresponding to the East of Europe and Russia,<br />
involving platforms and geo-dynamically active regions.<br />
Considering 63 km as the suggested depth of the asthenosphere<br />
in Seymour Island (also suggested by the MTS<br />
performed in James Ross Island), a heat flow of 78<br />
mWlm2 is estimated in this island (lpcallcm2sec. 42<br />
mWlm2). If the magma chamber, suggested by the MT<br />
study, really exists in James Ross Island its presence will<br />
increase the heat flow at surface. In fact, considering the<br />
top of this chamber at 6.8 km depth - according to MT results<br />
- a heat flow of 148 mWlm2 is estimated using the<br />
expressions mentioned. The heat flow values, thus estimated,<br />
should only be considered as approximations of<br />
the steady geothermal heat flows corresponding to these<br />
regions. It is important to mention the presence of a rift at<br />
the Bransfield Strait located <strong>20</strong>0 km to northwest from<br />
Seymour Island, where heat flows higher than 2<strong>20</strong><br />
mWlm2 were measured.<br />
In order to estimate the geothermal gradient in the<br />
permafrost of Seymour and James Ross Islands, a thermal<br />
conductivity obtained in the laboratory (I.883 Wlm OK)<br />
was used corresponding to two samples of frozen sands<br />
with about 87-94% of water. In this way a geothermal gradient<br />
of 0.04 "Clm is estimated for Seymour Island and<br />
0.079 "Clm for James Ross Island. Taking into account<br />
the MAAT for both islands: -9.4 "C for Seymour Island at<br />
<strong>20</strong>0 m a.s.l., (where the VES study gave <strong>20</strong>0 m of frozen<br />
ground thickness) and -5.0 "C for James Ross Island at<br />
about 35 m a.s.1. (where the VES study gave 40 to 45 m of<br />
frozen ground thickness), we obtain 235 m and 63 m as<br />
the estimated permafrost thickness corresponding to<br />
Seymour and James Ross Islands respectively.<br />
Borrotta, E. et.al.: Permafrost thickness in Antarctica usirrg.,<br />
Because of the high conductivity below the frozen<br />
ground put in evidence at both islands by MTS, the presence<br />
of a cryopeg at the base of the frozen ground is very<br />
probable. Therefore, the estimations of the permafrost<br />
thickness made in these islands (235 and 63 m) are consistent<br />
with the thickness of frozen grounds obtained by<br />
VES (<strong>20</strong>0 and 45 m). The differences could be attributed<br />
to a cryopeg.<br />
Although the permafrost thickness in equilibrium, thus<br />
estimated, may contain uncertainties, its comparison with<br />
the frozen ground thickness - corresponding to both<br />
studied islands - seems to suggest a state approaching<br />
equilibrium. This result means that the soil would have<br />
been uncovered by ice and with approximately stable climatic<br />
and geological conditions for a period of time long<br />
enough to reach this state.<br />
REFERENCES<br />
Adam, A. 1976. (ed.). Geoelectric and Geothermal Studies (east -<br />
central Europe, Soviet - Asia). KAPG Geophysical Monograph.<br />
Akademiai Kiado, Budapest.<br />
Adam, A. 1978. Geothermal effects in the formation of electrically<br />
conducting zones and temperature distribution in the Earth. Phys.<br />
Earth Planet. Inter. 17: 21-28.<br />
Del Valle, R.A., Demichelli, J., Febrer, J.M., Fournier, H.G., Gasco,<br />
J.C., Irigoin, H., Keller, M., Pomposiello, M.C. 1982. Resultats de<br />
deux campagnes magnktotelluriques faites dans le Nord la<br />
PBninsule Antarctique en 1979 et 1980. IX e Reun. Ann. Sci. de la<br />
Terre, Paris, SOC. Geol. Fr. ... dit., Paris.<br />
Fournier, H., Keller, M., Demichelli, J., Irigoin, H. 1980. Prospeccion<br />
magnetotelurica en la isla Vicecomodoro Marambio, Antartida.<br />
Direc. Nac. del Antartico, Inst. Antart. Argentino, Buenos Aires,<br />
Contribucion 235: 38-43.<br />
Fukuda, M., Strelin, J., Shimokawa, K., Takahashi, N., Sone, T.,<br />
Trombotto, D. 1992. Permafrost occurrence of Seymour Island<br />
and James Ross Island, Antarctic peninsula region. In: Y. Yoshida<br />
et al. (eds.), Recent Progress in Antarctic Earth Science: 745-750.<br />
12
Living microorganisms in Siberian permafrost and gas emission at low<br />
temperatures<br />
A. Brouchkov, M .Fukuda, K. Asano, M. Tanaka and F. Tomifa<br />
Hokkaido University, Sapporo, Japan<br />
Permafrost is a source of greenhouse gases. Thus thawing<br />
of the frozen soils affects conditions of the global carbon cycle.<br />
Organic material in the thawing permafrost decays<br />
quickly, releasing carbon dioxide and methane in the process.<br />
Long-term soil incubation experiments in flasks containing<br />
soil samples and nitrogen in the air have shown a slow pre<br />
duction of methane in different frozen soils at -5°C. Soils<br />
wereover-saturatedwithwater.Therewasanincrease in<br />
methane content in the air of the flasks, especially in the first<br />
<strong>20</strong>-50 days of the experiments (Brouchkov and Fukuda<br />
<strong>20</strong>02). The change in methane content occurred according b<br />
the logarithmical law insamplesofmodemsoilsfromYakutsk,<br />
Russia and Hokkaido (Tomakomai), Japan; the rates<br />
of methane production decrease with time (Table 1).<br />
world for many years, they might be still active. Techniques<br />
for sampling in permafrost for microbiological studies are not<br />
well-established. The requirement is for biological cleanliness<br />
and absolute avoidance of contamination while the sample is<br />
brought to the surface and during subsequent handling of the<br />
sample. Fieldwork done in <strong>20</strong>01-<strong>20</strong>02 in Yakutsk, Eastern<br />
Siberia, has shown that the permafrost at temperature -5°C of<br />
contains living fungi identified as Penidlium echinulafum<br />
(Figure 1).<br />
Table 1. Change of methane content (ppmv) in the air of the flasks<br />
during an incubation experiment at -5%<br />
Days Tomakomai soil Yakutsk soil<br />
0 1.09 0.85<br />
22 1.80 2.00<br />
50 1.89 2.07<br />
390 3.15 3.10<br />
Micro-organisms could be responsible for microbial methane<br />
formation in permafrost: according to recent studies,<br />
methane and gas hydrates are widely distributed in there and<br />
appear to be of a biogenic origin. Colonies of microorganisms<br />
were discovered that survived the incubation pe<br />
riod lasting longer than 1 year -5OCc; at they were able to undergo<br />
anaerobic growth at room temperature in GYP me<br />
dium; 16s rDNA was amplified by PCR.<br />
However, calculations of the amount of methane prcduced<br />
in permafrost based on short-term experiments are difficult.<br />
The production could be discontinued under certain conditions.<br />
The temperature is not stable: it could be suggested<br />
that methane production increases at higher temperatures. The<br />
type of soil, organic material and water content are also essential<br />
to microbial activity. In spite of the obvious possibility<br />
of methane production at low temperatures, long-term forecast<br />
of methane content in frozen soils is problematic.<br />
Permafrost soils contain micro-organisms (Friedmann<br />
1994, Gilichinsky and Wagener 1995); isolated from the<br />
The genera Penicillium are widespread and found in soil.<br />
Fungi are responsible for the biodegradation of organic material;<br />
this could be considered as one of their important practical<br />
applications. Some fungal species grow at low temperatures,<br />
below 0°C. However, there is no information<br />
concerning the fungi which may grow in underground perma-<br />
frost at temperatures below 0°C during a long time throughout<br />
their life cycle.<br />
The identification of permafrost fungi as Penici//ium<br />
13<br />
echinulafum was based on morphological characteristics and<br />
a nucleotide sequence analysis of enzymatically amplified
Brouchkov, A, et al, : Microorganisms in Siberian permafrost, gas emission ...<br />
18s rDNA, internal transcribed spacer (ITS) region including<br />
5.8s rDNA and DllD2 at the 5' end of the large subunit (26s)<br />
rDNA. The fungi were able to grow at positive and negative<br />
temperatures. The optimum temperature for the growth of P.<br />
echjnulaturn was estimated as I5-2OoC, yet they were able<br />
to grow at the negative temperature of -5°C (Figure 1). Fungi<br />
colonies on MEA grow rather rapidly, attaining a diameter d<br />
<strong>20</strong> to 35 mm in 7 to 9 days at <strong>20</strong>°C.<br />
organisms to those of known soil micro-organisms could re<br />
veal this mechanism.<br />
REFERENCES<br />
Ashcroft, F. <strong>20</strong>00. Life at the Extremes. HarperCollins.<br />
Brouchkov, A.V. and Fukuda, M. <strong>20</strong>02. Preliminary Measurements<br />
on Methane Content in Permafrost, Central Yakutia, and some<br />
Experimental Data. Permafrost and Periglacial Processes 3:<br />
187-197.<br />
Brouchkov, A.V. and Williams, P.J. <strong>20</strong>02. Could microorganisms in<br />
permafrost hold the secret of immortality What does it mean<br />
In: Contaminants in Freezing Ground. Collected <strong>Proceedings</strong> of<br />
2nd International Conference. Cambridge, England, Part 1: 49-<br />
56.<br />
Friedmann, E. I. 1994. Permafrost as microbial habitat. In: D. A Gilichinsky<br />
(ed.): Viable Microorganisms in Permafrost. Russian<br />
Academy of Sciences, Pushchino, Russia: 21-26.<br />
Gilichinsky, D. and Wagener, S. 1995. Microbial Life in Permafrost:<br />
A Historical Review. Permafrost and Periglacial Processes 6:<br />
243-<strong>25</strong>0.<br />
Williams, P.J. and Smith, M.W. 1989. The Frozen Earth, Cambridge:<br />
Cambridge University Press.<br />
Sampling of permafrost exposures in Aldan river valley<br />
(Mammoth Mountain, Eastem Siberia), in oneoftheoldest<br />
regions of permafrost on Earth (possibly aged about three million<br />
years) was done. Blocks of frozen soil and ice sized ap<br />
proximately <strong>20</strong>x<strong>20</strong>~<strong>20</strong> cm in size were cut from exposures<br />
and taken frozen to Sapporo. Frozen wood of the same age<br />
was also sampled.<br />
Micro-organisms have been discovered (Figure 2). They<br />
were able to grow at both aerobic and anaerobic conditions.<br />
For the sequence analysis DNA was extracted using the<br />
ISOPLANT set (Nippon Gene) and following the manufacturer's<br />
instructions. Preliminary results of sequencing and<br />
analysis of the evolutionary tree of the micro-organisms show<br />
that they are most related to Bacillus anthracis.<br />
Understanding these' permafrost organisms and their relationship<br />
to their cold environment, especially to unfrozen water<br />
common in frozen soils (Williams and Smith 1989), has<br />
immediate practical implications. The fact that viable bacteria<br />
occurring at significant depths in permafrost can be prompted<br />
to reproduce, presents imaginative possibilities, for example<br />
for the decontamination of buried contaminants. The living micro-organisms<br />
in permafrost, in common with other extremophiles<br />
(Ashcroft <strong>20</strong>00), apparently have special mechanisms<br />
of repair of cell structures, necessary for their survival<br />
(Brouchkov and Williams <strong>20</strong>02). A comparative study d<br />
structure and biochemical features of permafrost's micro-<br />
14
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Natural risk evaluation of geocryological hazards (the Varandey Peninsula<br />
coast of the Barents Sea)<br />
1. V. Chekhina and F. M. Rivkin<br />
industrial and <strong>Research</strong> lnstifute for Engineering of Construction, Moscow, Russia<br />
INTRODUCTION<br />
It is known that the intense technological impact on the Russian<br />
Arctic coast caused by industrial engineering results in<br />
the intensification of natural and origination of new geocryological<br />
processes. Therefore, prediction of development d<br />
geocryological processes is part and parcel of geotechnical<br />
survey.<br />
The intensity of geocryological processes can be estimated<br />
only on the basis of complex engineering geocryological<br />
studies. Therefore, the creation of information databases<br />
on the basis of engineering geocryological studies is an element<br />
of primary importance for solving the complex of prob<br />
lems of environmental management.<br />
In this connection, it becomes possible and necessary b<br />
automate construction of different engineering geocryological<br />
maps using geoinformation systems (GIs). This makes it<br />
possible not only to solve traditional problems of engineering<br />
survey but also to map natural risks and to assess the possibility<br />
of appearance of geocryological processes and their intensity.<br />
METHODS AND RESULTS.<br />
the engineering geocryological zoning, using the number d<br />
Results of the engineering geocryological studies performed exogenous processes and the degree of appearance risk d<br />
on the Varandey Peninsula were used as base information b geocryological hazards. Thefirstevaluationtablehasbeen<br />
estimate the intensity of geocryological processes. compiled using the total number of available hazardous ex-<br />
The electronic database of geocryological conditions was ogenous processes (frost heaving, soil settlement upon<br />
compiled on the basis of these studies. The database includes thawing, thermal erosion, landslides, and cryogenic appearsuch<br />
attributes as engineering geological regions, which were ances) and ignoring the appearance risk in the regions distindistinguished<br />
taking into account the geomorphological posi- guished. The second evaluation table reflects the total appeartion,<br />
genesis and lithology of soils. Moreover, physical ance risk of the processes in each of the region, additionally<br />
(moisture, ice content, density, etc.) and thermal (heat capac- including estimation of ice content and salinity of soils. The<br />
ity, thermal conductivity, etc.) properties of thawed and frozen total score of complexity of each geocryological region is<br />
soils were also taken into account. Each region distinguished<br />
was indexed depending on geocryological conditions.<br />
The database has a two-level architecture. Data obtained<br />
as a result of the field studies and laboratory determination d<br />
soil properties is combined at the first level. The second level<br />
includes generalized information, namely first-level data and<br />
information obtained from publications. Since the second-level<br />
database includes information from different sources, all input<br />
parameters are tested for mutual consistency.<br />
As a result, the compiled (summarized) database provides<br />
the complete set of properties necessary for analyzing<br />
the environmental quality in each of the distinguished regions.<br />
The structure of the obtained database is correlated with the<br />
scheme of zoning and is factually the legend to the map.<br />
Natural risks of such geocryological hazards as frost<br />
heaving, soil settlement upon thawing, thermal erosion, landslides,<br />
and cryogenic appearances were evaluated based on<br />
the compiled database. Ice content and salinity of soils were<br />
also estimated in order to assess the complexity of the area.<br />
The rule base (analytic relationships for scores) and pro.<br />
gram module for calculating scores were created in order b<br />
evaluate natural risks of geocryological hazards. The module<br />
for calculating scores from analytic relationships makes it<br />
possible to determine an appearance risk of exogenous gemryological<br />
processes (Dan'ko, 1982; Lovchuk, 1990). A<br />
certain score is assigned to each cryogenic processes depending<br />
on the obtained index value. The score system is<br />
based on risk evaluations of each geocryological process<br />
(Armand, 1975).<br />
Thus, two evaluation tables were obtained. These tables<br />
make it possible to differentiate regions, distinguished during<br />
given for the most hazardous process, assuming that effect its<br />
will be higher than the effect of the remaining processes.<br />
As a result, all engineering geocryological regions were<br />
differentiated with respect to the degree of the appearance risk<br />
of hazardous processes in the following way: most<br />
hazardous, medium hazardous, and least hazardous. Data d<br />
both summary tables were reduced to the electronic engin-<br />
15
eering geocryological map, which is directly related to the<br />
database.<br />
The electronic database of the key site of the Varandey Peninsula,<br />
which includes the engineering geological characterization<br />
of the natural environment, was created as a result d<br />
the performed studies. The algorithm was constructed for<br />
evaluating natural risks of appearance of such geocryological<br />
hazards as frost heaving, soil settlement upon thawing, thermal<br />
erosion, landslides, and cryogenic appearances and for<br />
estimating ice content and salinity of soils.<br />
Two evaluation maps of the key site of the Varandey Peninsula<br />
were constructed on the basis of theelaboratedtechnique:<br />
- Map of zoning with respect to the appearance risk of the<br />
most hazardous geocryological process (Fig. 1);<br />
- Map of distribution of exogenous geocryological processes<br />
(Fig. 2).<br />
CONCLUSION<br />
The system for interpreting results of geotechnical survey,<br />
using electronic maps, electronic databases, and program<br />
modules, has been created as a result of the performed studies<br />
in order to evaluate the distribution of exogenous processes.<br />
Theelaborated system can be used to map natural<br />
risks and to construct prediction maps of development of hazardous<br />
exogenous processes.<br />
ACKNOWLEDGEMENT<br />
This work is supported by Russian Coast Protection Program<br />
and INTAS grant 2332.<br />
REFERENCES<br />
Armand, D.L. 1975. Landscape Science, Moscow: Mysl'<br />
Dan'ko, V.K. 1982. Regularities of Development of Thermal Erosion<br />
in Northwestern Siberia, Cand. Sci. (Geol.-Min.) Dissertation,<br />
Moscow.<br />
Lovchuk, V.V. 1990. Cryomorphogenesis of Subarctic Lowlands in<br />
Western Siberia: Backgrounds of a Topography Cryomorphological<br />
Analysis, Methods of Engineering Geocryological<br />
Survey, Moscow: VSEGINGEO.<br />
Figure 1. Map of zoning of the key site of the Varandey Peninsula<br />
with respect to the appearance risk of the most hazardous geocry-<br />
ological process.<br />
1 - most hazardous: 2 - medium hazardous; 3 I<br />
least hazardous.<br />
Figure 2. Map of distribution of exogenous geocryological processes<br />
at the key site of the Varandey Peninsula.<br />
1 - most hazardous; 2 - medium hazardous; 3 - least hazardous.<br />
16
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available information<br />
AIGEO working group “Past distribution of the periglacial environment in<br />
the Italian mountain chains: reconstructions based on the study of relict<br />
forms”<br />
Alessandro Chelli, *M. Pappalardo and A. Ribolini<br />
Dipartimenfo di Scienze della Terra, Universifa di Parma, Parma, lfaly<br />
* Dipartimenfo di Scienze della Terra, Universifa di Pisa, Pisa, lfaly<br />
A working group within the Italian Association of Geomor- periglacial landforms in their relict state (e.g., rock glaciers<br />
phologists started its activity at the end of <strong>20</strong>02. A number and wedge casts) for Quaternary paleoclimatic reconof<br />
Italian researchers, engaged in studies on the perigla- structions in Italian low altitude mountain environments.<br />
cia1 environment and often dealing with relict forms, decided<br />
to get together in order to actively compare their<br />
d 15 I<br />
7<br />
study objects and methods.<br />
In Italy, landforms related to periglacial processes have<br />
been recognized for a long time (Capello 1960) and permafrost<br />
distribution in the present and throughout the<br />
Quaternary has recently been reconstructed by means of<br />
typical landforms (e.g. rock glaciers; Dramis and Gug-<br />
/<br />
lielmin 1999). However, many other landforms, e.g., different<br />
types of inactive scree, common mainly in the Apennine<br />
chain and in Sardinia, were identified in the past<br />
and considered as “periglacial”. Although displaying different<br />
characteristics with regard to parent material, texture,<br />
form, position and extension, they have some fea-<br />
1<br />
tures in common: they are located on moderately steep<br />
slopes, are made of a coarse debris that may be mixed<br />
with fine material and they have no sedimentary features<br />
typical of fluvial or gravitational landforms; in all cases<br />
they are not related to present-day geomorphic processes.<br />
In the literature they were classified mainly as eboulis ordonnees,<br />
greze lite6e and block streamslfields.<br />
The working group members intend to collect evidence<br />
i<br />
in order to clarify the true genetic environment of such<br />
landforms: whether they are really periglacial, or whether Fig, 1 Location of the study areas.<br />
just cryogenic weathering and frost shattering can be responsible<br />
for coarse sediment supply, in the absence of Study areas were established (Fig. 1) and a prelimieven<br />
sporadic permafrost. The possibility that other nary illustration of the deposits concerned permitted the<br />
weathering processes have been responsible for the for- members to compare the different typologies. Within the<br />
mation of such deposits, and that the action of gravity or category of block streamslfields study cases were serunning<br />
waters in the transport of weathered material has lected in Sardinia, Tuscany and Western Liguria. Sites<br />
until now been misunderstood, will be carefully examined. with slope deposits of likely cold climate environment were<br />
In fact, stratified scree is no longer strictly considered a enclosed both from the Northern Apennines (Parma and<br />
typical periglacial landform, as multiple processes can be Reggio Emilia sectors) and from the Central (Abruzzo and<br />
evoked for its genesis (Van Steijn et al. 1995). Moreover, Molise) and Southern Apennines (Cilento), as well as from<br />
block streams, presently being studied as active forms in Sardinia. Active landforms testifying to the present-day<br />
extreme environments (very cold and arid), were recog- transition from glacial to periglacial environment in the<br />
nized to be convergent, in their relict form, with other, non Gran Sasso area (Central Apennines), especially lobated<br />
periglacial, types of landforms (Boelhouvers et al. <strong>20</strong>02). debris indicating permafrost creeping processes, will be<br />
Another aim of the working group is to employ typical included among the investigation objects. Their analysis<br />
17
can be important, in fact, for making a comparison with the<br />
studied relict forms.<br />
The first goal of the working group will be the creation<br />
of a work sheet in order to perform a description of the<br />
different deposits with identical methods. Micromorphological<br />
techniques and image analysis, successfully<br />
employed by some of the members on single deposits, will<br />
be extended wherever possible. Further work will focus on<br />
the chronological aspect of the matter and the necessity of<br />
finding a suitable dating method that permits comparison<br />
of the ages of the studied bodies.<br />
After four years’ activity, the goal of the working group<br />
is to define the role that frost weathering had in sediment<br />
supply, the role of frost creep and gelifluction processes in<br />
the material accumulation, and the relationship between<br />
the climate under which such landforms were created and<br />
the distribution of mountain permafrost.<br />
Working Group Members: A. Bini (Universita di Milano), N.<br />
Casarosa (Provincia di Pisa), A. Chelli (Universifa di<br />
Parma), M. Coltorfi (Universifa di Siena), P. Farabollini<br />
(Universifa di Camerino), M. Firpo (Universiti di Eenova),<br />
M. Guglielmin (ARPA Lombardia), E. Miccadei (Universifa<br />
di Chieti), M. Pappalardo (Universifa di Pisa), M. Pecci<br />
(ISPESL Roma), T. Piacenfini (Universitd di Chiefi), F.<br />
Scarciglia (Universifi della Calabria), C. Queirolo (Universifa<br />
di Genova), R. Rafi (Universita di Roma “La Sapienza’J,<br />
A. Ribolini (Universifb di Pisa), s. Sias (Universita di<br />
Sassari), C. Tellini (Universifa di Parma).<br />
The abstract volume of the Working Group firsf meeting is<br />
available upon request.<br />
Chelii, A. et al.: AIGEO, periglacial environment. reconstructions, relict form<br />
REFERENCES<br />
Boelhouvers J., Holness S., Meiklejohn I. and Sumner P. <strong>20</strong>02. Observations<br />
on a blockstream in the vicinity of Sani Pass, Lesotho<br />
highlands, Southern Africa. Permafrost and Periglacial Processes<br />
13(4): <strong>25</strong>1 -<strong>25</strong>7.<br />
Capello C.F. 1960. Terminologia e sistematica dei fenomeni dovuti al<br />
gelo discontinuo. G. Chiappichelli, Torino.<br />
Dramis F. and Guglielmin M. 1999. Permafrost investigations in the<br />
Italian mountains: the state of the art. In Paepe R. & Melnikov V.<br />
(eds.), Permafrost response on economic development, environmental<br />
security and natural resources, Kluver Ac. Publishers:<br />
<strong>25</strong>9-273<br />
Van Steijn H., Bertran P., Francou E., HBtu B. and Texier J.P. 1995.<br />
Models for the genetic and environmental interpretation of stratified<br />
slope deposits: review. Permafrost and Periglacial Processes<br />
6: 1<strong>25</strong>-146.<br />
18
8th ~ n ~ ~ Conference ~ ~ ~ on Permafrost ~ ~ i Extended ~ n ~ Abstracts l on Current <strong>Research</strong> and Newly Available i ~ f ~ r ~ ~ ~ i o ~<br />
Active layer dynamics in Greenland, Svalbard and Sweden<br />
H. H. Christiansen, *J. H. Akerman and **J. Repelewska-Pekalowa<br />
lnstitute of Geography, University of Oslo, Norway<br />
* lnstitute of Physical Geography and Ecosystems Analysis, University of Lund, Sweden<br />
"*Department of Geomorphology, University of Maria Curie-Sklodowska, Poland<br />
INTRODUCTION<br />
Four sites with the longest active layer monitoring records<br />
in high arctic NE Greenland and Svalbard, and in alpine<br />
Sweden are compared. In the Greenland Sea, deep-water<br />
formation occurs bringing warm surface water from the<br />
south. This provides a relatively mild climate at high latitudes.<br />
Meteorological data and GCMs all suggest that climatic<br />
changes at high latitudes are amplified as compared<br />
to lower latitudes. Therefore, active layer monitoring and<br />
its relationship to climatic variations in Greenland, Svalbard<br />
and northern Scandinavia are of high relevance for<br />
monitoring and study.<br />
The active layer monitoring sites are located at Abisko,<br />
northern Sweden, 68ON 19OE, at Calypsostranda, 77ON<br />
15OE, and at Kapp Linne, 78ON 13OE, both at Svalbard,<br />
and at Zackenberg, 74ON 2I0W, NE Greenland. We discuss<br />
intersite and interannual variations in active layer<br />
thickness and its potential relationship to summer air temperature<br />
represented by thawing degree days.<br />
This abstract was developed as a result of the CALM<br />
synthesis workshop held in November <strong>20</strong>02 at the University<br />
of Delaware, USA.<br />
ground moraine at 36 m a.s.1 (Christiansen 1999). All sites<br />
are located in open flat landscapes with average landscape<br />
snow cover thickness and duration conditions. Meteorological<br />
stations operated since grid establishment at<br />
the Zackenberg Station <strong>25</strong> m from the Zackenberg grid;<br />
until 1976 and again from 1996 at lsfjord 0.5-5 km from<br />
the Kapp Linne grids; at the Hornsund Station, which is 68<br />
km south of Calypsostranda; and at the Abisko Station 3-<br />
50 km from the Abisko grids. In the last <strong>20</strong>-<strong>25</strong> years there<br />
has been an increase (1-2 OC) in summer temperatures<br />
from the Abisko and Longyearbyen, Svalbard stations.<br />
This warming has, however, not occurred in Greenland.<br />
DATA AND DISCUSSION<br />
SITE DESCRIPTIONS<br />
At Abisko, data on active layer thickness were obtained<br />
beginning in 1978 at eight valley palsa bog grids at 380-<br />
480 m a.s.l., and at three intermontane depression grids at<br />
850-950 m a.s.l., each with 121 measuring points (Brown<br />
et al., <strong>20</strong>00). At Kapp Linne, monitoring was initiated in<br />
1972 at ten I00 x I00 m grids on raised marine terraces<br />
at 4-59 m a.s.l., nine on dry mineral soils and one in a<br />
peat-covered area, each with <strong>25</strong> randomly sampled<br />
measuring points until 1994, and after that with 121 points<br />
(Brown et al. <strong>20</strong>00). At Calypsostranda, active layer<br />
monitoring started in 1986 at 23 individual points on raised<br />
marine terraces at 2-40 m a.s.1. These points adequately<br />
represent the average landscape active layer thickness<br />
based on a total of 300 measuring points (Repelewska-<br />
Pekalowa <strong>20</strong>02). At Zackenberg, a grid of I00 x 100 m,<br />
with 121 measuring points was established in 1996 on<br />
1975 1980 1986 1990 21896<br />
m<br />
Figure 1. Active layer thickness (cm) at Abisko in Sweden, Kapp Lime<br />
and Calypsostranda at Svalbard and at Zackenberg in NE Greenland.<br />
No data was collected from the Calypsostranda sites in 1994, 1997<br />
and 1999.<br />
At the Abisko valley site, active layer thickness was<br />
stable around 55 cm from 1978 to 1995, and increased to<br />
around 80 cm since then (Fig. 1). In the Abisko Mountains<br />
the active layer thickness varied from 85 to 1<strong>20</strong> cm during<br />
the measuring period, with the largest values in the last<br />
five years. The Kapp Linne sites show decreasing active<br />
layer thickness until 1987 and then increasing, with the dry<br />
sites varying around 75 to 110 cm, and the wet sites from<br />
19
<strong>25</strong> to 95 cm. The thickest active layer in the region occurs<br />
at Calypsostranda, varying only from 115 to 140 cm for<br />
both wet and dry sites. There is no trend in active layer<br />
thickness at Calypsostranda. At Zackenberg the active<br />
layer is only 60 to 70 cm thick. This is the smallest value in<br />
mineral soil, and this site, with its relatively short record,<br />
also has the smallest interannual variation of the four<br />
sites.<br />
A common pattern of interannual active layer thickness<br />
variation exists from 1997 to <strong>20</strong>00 (Fig. 1). The Calypsostranda<br />
sites were not monitored during all these years, but<br />
data from the monitored years fit the pattern. In 1980 and<br />
1988 the Kapp Linne and Abisko sites have peaks in the<br />
active layer thickness, but this was not seen at Calypsostranda<br />
in 1988. In 1989 the active layer thickness was reduced<br />
for all Kapp Linne and Abisko sites, but thickest<br />
during the recording period at Calypsostranda, while in<br />
1990 all Kapp Linne and Abisko sites peaked, when Calypsostranda<br />
sites were small.<br />
= 0 (n=14)<br />
Calypsostranda wet<br />
rz = 0.04 (n=14)<br />
v v<br />
V<br />
v It 0.31 (n=30)<br />
f ~ app Linne wet<br />
P 0.59 (n=29)<br />
Christiansen, H.H. et al,: Active layer dynamics in Greenland, Svalbard ...<br />
Abisko mountain<br />
P = 0.47 (n=26)<br />
140<br />
1<strong>20</strong><br />
1 DO<br />
shallow Zackenberg active layer in mineral soil of 60-70<br />
cm is comparable to the peat sites in Svalbard and northern<br />
Sweden. The shallow Greenlandic active layer reflects<br />
the East Greenland Current carrying arctic water masses<br />
south along East Greenland, causing relatively colder<br />
summers here than at Svalbard and in northern Scandinavia.<br />
On Svalbard, Calypsostranda is located only 58 km<br />
south of Kapp Linne, and both sites are located on raised<br />
marine terraces of the west coast. However, the active<br />
layer at the Calypsostranda sites is significantly thicker<br />
than even at the Kapp Linne mineral sites. The Hornsund<br />
and Kapp Linne summer air temperatures are within the<br />
same range, have equal overall interannual variations, but<br />
with the Hornsund values slightly lower than the Kapp<br />
Linne, particularly since 1995. During the active layer<br />
monitoring period, summer air temperatures have been<br />
rising mainly since 1998 at Kapp Linne, but not as significantly<br />
at Hornsund. Whereas the active layer increased at<br />
Kapp Linne already from 1988, there is no increase in active<br />
layer thickness at Calypsostranda. This combination<br />
of information indicates that local conditions at both sites<br />
have a larger control on active layer thickness than summer<br />
air temperature. This is in agreement with what has<br />
been found for other CALM sites (Brown et al. <strong>20</strong>00).<br />
None of the locations have a significant correlation between<br />
active layer thickness and DDT, stressing that the<br />
control by local factors are as strong as the influence from<br />
summer air temperature. It is also possible that other meteorological<br />
conditions could influence the active layer<br />
thickness as much as the summer air temperature. This<br />
remains to be investigated. The interannual active layer<br />
thickness variation pattern from 1997 to <strong>20</strong>00 for all sites<br />
show similar response, presumably controlled by some<br />
regional effect, however, not by the air temperature alone.<br />
REFERENCES<br />
Figure 2. Active layer depth and the square root of thawing degree<br />
days of the air temperature at the closest meteorological stations at<br />
the four sites.<br />
To investigate the relationship between active layer<br />
thickness and summer air temperature, thawing degree<br />
days (DDT) were calculated using the closest meteorological<br />
station for each four investigated sites in the years<br />
with active layer measurements (Fig. 2). There is a positive<br />
correlation for the Kapp Linne and Abisko sites, and<br />
particularly for the peat bog sites, but none for the Zackenberg<br />
and Calypsostranda sites.<br />
The main intersite difference in active layer thickness<br />
between Greenland, Svalbard and Sweden primarily reflects<br />
climatic and material differences. Both peat sites at<br />
Abisko and at Kapp Linne have a generally shallow, <strong>20</strong>-70<br />
cm active layer, while the mineral soils here have thicker<br />
active layers of 80-1<strong>20</strong> cm. The active layer in the peat<br />
site at Kapp Linne is thinner than the Abisko alpine peat<br />
sites, while the active layer in the Abisko Mountains is<br />
slightly thicker than at the mineral sites at Kapp Linne. The<br />
<strong>20</strong><br />
Brown, J., Hinkel. K.M., and Nelson, F.E. <strong>20</strong>00. The Circumpolar Active<br />
Layer Monitoring (CALM) Program: <strong>Research</strong> Designs and<br />
Initial Results. Polar Geography, 24: 165-<strong>25</strong>8.<br />
Christiansen, H.H. 1999. Active layer monitoring in two Greenlandic<br />
permafrost areas: Zackenberg and Disko Island. Danish Journal<br />
of Geography, 99: 117-l2lI<br />
Repelewska-Pekalowa, J. <strong>20</strong>02. CALM - Site P1- Calypsostranda.7.<br />
Akerman, J.H. 1980. Studies on Periglacial Geomorphology in West<br />
Spitsbergen. Meddelanden frAn Lunds Universitets Geografiska<br />
Institution Avhandlingar LXXXIX.
8th International Conference an ~~~~~~~~~~<br />
Extended Abstracts an Current <strong>Research</strong> and Newly Available Information<br />
Could the current warming endanger the status of the frozen ground<br />
regions of Eurasia<br />
S. M. Chudinova, S.S. Bykhovets, V.A. Sorokovikov, D.A. Gilichinsky, *T-J. Zhang and R. G. Barry<br />
Soil Cryology laboratory, lnsfitute of Physicochemical and Biological Problems in Soil Science, Russian Academy of Sciences,<br />
Pushchino, Moskow oblast, Russia<br />
*Natinal Snow and Ice Data Center, University of Colorado, Boulder, USA<br />
INTRODUCTION<br />
The current rise in air temperature that started in the second<br />
part of the 1960s in the northern territory of Russia<br />
takes place predominantly in the cold period (Watson et<br />
al. 1998). However, a number of impacts induced by climate<br />
warming, such as reduction of the permafrost area<br />
and changes in thickness of the active layer, etc., are determined<br />
mainly by changes in ground temperature during<br />
the warm period. Changes in ground temperature reflect<br />
not only changes in air temperature but also combined<br />
impacts from many other climatic factors. The seasonal<br />
trends of ground temperature do not necessarily correspond<br />
with the trends observed for air temperature (Gilichinsky<br />
et al. 1998). Zhang et al. (<strong>20</strong>01) found that abundant<br />
rainfall under global warming may lead to a decrease<br />
in summer ground temperature. Therefore, the actual<br />
danger of an air temperature increase can be assessed<br />
only on the basis of direct determination of changes in<br />
ground temperature. This work investigates the vast regions<br />
of Eurasia, where the current warming represents a<br />
potential danger.<br />
METHODS<br />
The investigation was made for the period 1969-1990, for<br />
which the most comprehensive data on ground temperature<br />
are available. Linear trends were used to estimate<br />
tendencies in the time series of ground and air temperature.<br />
Analysis was performed for the mean annual ground<br />
temperature, mean annual and monthly air temperature,<br />
as well as ground temperature averaged over the summer<br />
(June-August), winter (December-February), spring<br />
(March-May), and autumn (September- November) periods.<br />
Ground temperature was measured at depths of 40,<br />
160 and 3<strong>20</strong> cm. Best-fit linear trends were calculated for<br />
individual stations. Regions with different responses of<br />
seasonal ground temperature to the air warming were<br />
thereby delimited (Fig. 1). Then, time series of average<br />
trends of ground temperature were calculated for each region<br />
by the method of polygon area averaging. Trends of<br />
ground temperature equal to 0.04"Clyear and above are<br />
significant at the 0.05 level.<br />
RESULTS<br />
A correspondence between changes in the time series of<br />
annual air and ground temperature has been identified<br />
down to a depth of 3<strong>20</strong> cm for the territory under investigation.<br />
Maximum trends of air temperature, and hence<br />
maximum trends of annual ground warming down to 3<strong>20</strong><br />
cm are observed for the territories with permafrost and<br />
seasonally frozen ground in the West Siberian Plain (region<br />
IV), the central Siberian plateau (region VI), and the<br />
mountains of Transbaikalia (region VIII). Here, trends of<br />
ground temperature at the three depths analyzed average<br />
0.07-0.06, 0.05-0.08, 0.05-0.06"Clyear, respectively. Annual<br />
ground warming is practically absent in the zones of<br />
discontinuous permafrost in Pribaikal'e (region VII) (0.01-<br />
0.03"Clyear) and the seasonally frozen grounds of Far<br />
East (region X) (O.O-O.O2"Clyear), and over vast territory<br />
occupied by continuous permafrost to the east of the Lena<br />
River (region XI) (-0.01-0.O"Clyear). This reflects an absence<br />
of any significant increase in the annual air temperature<br />
time series in these regions.<br />
The relationship between mean seasonal time series of<br />
air and ground temperature is more complicated. In the<br />
taiga zone of the Russian Plain (regions I and II), changes<br />
in ground temperature are controlled mainly by air temperatures<br />
of the warm period (Chudinova et al. <strong>20</strong>01). In<br />
the northern and middle taiga zones (region I), in spite of<br />
the predominant increase in winter air temperature, the<br />
major rise in ground temperature is observed in the warm<br />
period (Fig. lb), which is dictated by the strong rise in air.<br />
The absence of summer air warming in the southern taiga<br />
(region II) results in the absence of a ground temperature<br />
increase during the whole year.<br />
21
In the Ural mountains (region Ill) and West Siberian<br />
Plain, a strong rise in air temperature in both the warm<br />
and the cold periods results in large positive trends in both<br />
winter and summer (0.08-0.1 O'Clyear) ground temperature<br />
(0.07-0.10 in the upper layers and 0.04-0.08'Clyear<br />
in the lowest one). In the vast permafrost zone of East Siberia<br />
(regions VI, VII, VIII, and XI), trends in the ground<br />
temperature time series are determined mainly by<br />
changes in winter air temperature. The effect of changes<br />
in summer air temperature extends only down to 40 cm.<br />
Therefore, in accordance with air temperatures trends in<br />
this region, ground temperature increases in the cold period<br />
(autumn-winter-spring) and remains unchanged, or<br />
decreases, in summer (Fig. 1). The largest significant rise<br />
in ground temperature in the cold period (winter-spring),<br />
down to a depth of 3<strong>20</strong> cm, characterizes the central Siberian<br />
Plateau and mountains of Transbaikalia, averaging<br />
0.10-0.08"Clyear. The summer trend of ground temperature<br />
in these regions is zero or negative to a depth of 40<br />
cm. Below, due to winter warming, trends of ground temperature<br />
become positive and reach 0.05-0.06"Clyear.<br />
Trends at individual stations can differ strongly from the<br />
average estimates. Therefore, we attempted to ascertain<br />
whether this increase takes place at stations where permafrost<br />
exists from a depth of 160-3<strong>20</strong> cm. Our analysis<br />
shows the absence of significant trends in mean monthly<br />
ground temperature, both at the boundary of the thaw<br />
layer and in the permafrost. The maximum decrease in<br />
summer ground temperature is observed over the vast<br />
discontinuous permafrost zone in region XI. A strong rise<br />
in air temperature in February and March cannot offset<br />
this cooling with depth. Thus, changes in permafrost<br />
status in this region have not occurred.<br />
P<br />
-0.03 0.0 0.03 0.06 0.09<br />
Chudinova, S. M. et at.: Current warming, frozen ground, Eurasia..<br />
CONCLUSION<br />
Despite a significant rise in winter and annual ground<br />
temperature in the vast territory of East Siberia, no significant<br />
effect on the permafrost status is apparent. Based on<br />
the results obtained, attention should be focused on the<br />
West Siberian Plain, where extensive territories are occupied<br />
by permafrost and peatland. Changes in the temperature<br />
regime of these ecosystems can have both<br />
negative and positive consequences (for example, for agriculture).<br />
The West Siberia Plain is the region of Russia<br />
most sensitive to global air change (Anisimov <strong>20</strong>01).<br />
Therefore, changes observed in the ecosystems (especially<br />
peatlands) of West Siberia may be of interest for<br />
other regions of the world where similar changes are taking<br />
place.<br />
ACKNOWLEDGEMENT<br />
This work was supported by the Russia Foundation for<br />
Basic <strong>Research</strong> (02-05-64597) and in part by NSF-OPP<br />
(9907541& 0229766).<br />
REFERENCES<br />
Anisimov, O.A. <strong>20</strong>01. Predicting patterns of near-surface air temperature<br />
using empirical data. Climatic Change 50: 297-315.<br />
Chudinova, S.M., Bykhovets, S.S., Fedorov-Davydov, D.G., et al.<br />
<strong>20</strong>01. Response of temperature regime of Russian North to climate<br />
changes during second half of <strong>20</strong>th century. Earth<br />
Cryosphere 5(3): 63-69 (in Russian).<br />
Watson, R.T. Zinyowera, M.C. and Moss, R.H. (eds). 1998. The Regional<br />
Impacts of Climate Change. An Assessment of Vulnerability.<br />
IPCC, Cambridge University Press.<br />
Zhang, T, Barry R.G., Gilichinsky, D.A., Bykhovets, S.S., et al. <strong>20</strong>01.<br />
An amplified signal of climatic change in ground temperatures<br />
during the last century at Irkutsk, Russia. Climatic Change 49: 41-<br />
76.<br />
Gilichinsky, D.A., Barry, R.G., Bykhovets, S.S. et al. 1998. A century<br />
of temperature observations of soil climate: methods of analysis<br />
and long-term trends. In A.G. Lewkowicz & M. Allard (eds), Permafrost;<br />
Proc. the 7th intern. conf., Yellowknife, 23-27 June, 1998,<br />
Universitk Laval: 313-17.<br />
Figure 1. Regions (I-XI) with different linear trends of ground temperature<br />
at a depth of 40 cm for (a) winter and (b) summer.
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Snow effects on recent shifts (1998-<strong>20</strong>02) in mean annual<br />
temperature at alpine permafrost sites in the western<br />
R. Delaloye and M. Monbaron<br />
ground surface<br />
Swiss Alps<br />
Department of Geosciences, Geography, University of Fribourg, <strong>Switzerland</strong><br />
INTRODUCTION<br />
The ground temperatures at several alpine permafrost sites<br />
have been monitored for the last five years. The data analysis<br />
highlights the effects of both duration and thickness of the<br />
snow cover on the mean annual ground surface temperature<br />
(MAGST).<br />
The calculation of MAGST is frequently used for estimating<br />
the occurrence of permafrost. Nevertheless, as weather<br />
conditions may dramatically fluctuate from one year to another,<br />
MAGST is susceptible to large interannual changes,<br />
which can lead to significant misinterpretation of the data. The<br />
snow conditions represent a key factor governing the shortterm<br />
variations in MAGST. Concerning the amplitude d<br />
these variations, Vonder Miihll et al. (1998) relate, for instance,<br />
a MAGST drop of more than 2°C at a depth of 0.6 m<br />
in a borehole over a one-year interval. However, the spatial<br />
variability of the temporal evolution of MAGST still remains<br />
poorly documented. The present work reports observations<br />
made at both regional and local scales.<br />
REGIONAL SCALE<br />
Figure 1 provides a regional overview of the behaviour of<br />
MAGST. For each site, the values correspond to the average<br />
of all the UTL-I . A first observation is that the MAGST is<br />
relatively high for permafrost terrain, frequently above 0°C.<br />
This could be see as an effect of climate warming, with Ute<br />
potential for a rapid degradation of permafrost. More probably,<br />
however, these high temperatures indicate a "thermal offser<br />
between the surface and the top of the perennially frozen<br />
ground, a matter hat still requires further study of mountain<br />
permafrost.<br />
I I I I<br />
the<br />
SITES AND METHOD<br />
The analyzed data stem from five sites, each less than 1 km*<br />
-2.0<br />
wide, and located within a radius of <strong>20</strong> km in the western ! I I I I I<br />
1 1.38 l,l.9Y 1 1.30 1.1 31 l.l:oz Date<br />
Swiss Alps. The regional climatic conditions are transitional<br />
fd.rn.YY)<br />
-R chy-N-2800-(4) -Mllle-NE-2350-(7)<br />
between the humid northwestern fapde of the Alps (Mont- ..-... Ritord-W-2853-(17) -Aget-SE-2850-(1)<br />
Blanc), the southern quite mediterrannean side of the alpine -.-...- Ssnetsch-N.2350-r71<br />
range, and the drier inner parts of Valais. The Sanetsch site is Figure 1. Monthly running MAGST during the period 1998-<strong>20</strong>02.<br />
an exception as it is directly located on the northern and wet- For each site, the values correspond to the average of n UTL-1 and<br />
are placed at the end date of the annual interval. Legend: siteter<br />
border of the Alps.<br />
aspect-mean elevation-n. Black and white dots indicate the extreme<br />
Depending on the site, 4 to 22 minidataloggers (UTL-I) values of MAGST calculated for all sites together, the arrows show<br />
have recorded the temperatureat2-hour intervals, the first the date when they occurred.<br />
ones since October 1997. The annual series are adjusted on<br />
the zero curtain phases. The resulting accuracy of measure- Figure 1 also shows that the interannual evolution of the<br />
ment is +/-0.15"C. The UTL-1 have been positioned immedi- MAGST looks quite similar for every site. An increase in the<br />
ately below the ground surface in order to be protected from values from the end of 1999 until April <strong>20</strong>01 was followed by<br />
direct solar radiation, i.e., at a depth of between IOand 70 a strong decrease during the next 12 months. On average,<br />
cm, depending on the ground granulometry and on the defini- over the 41 available UTL-1, the drop in MAGST reached<br />
tion of the surface.<br />
1.4"C between April <strong>20</strong>01 and April <strong>20</strong>02. The maximum<br />
decrease obtained for a single datalogger was 3.I0C, the<br />
23
minimum only 0.3%. The shift in ground surface temperature<br />
was not accompanied by any significant change in mean annual<br />
air temperature (MAAT, Fig. 1). Indeed, interannual differences<br />
in snow covering caused this thermal drop.<br />
Important snowfalls occurred relatively early in autumn<br />
<strong>20</strong>00 and then protected the ground from winter cooling. The<br />
effect on MAGST was an increase towards the maximal<br />
values recorded during the five-year monitoring period. The<br />
spring of <strong>20</strong>01 was again characterized by subsequent<br />
snowfalls which caused a relative delay in the snowmelt in<br />
comparison with the previous year. In response, MAGST<br />
began to decrease. The autumn of <strong>20</strong>01 was not snowy and<br />
was followed by very cold weather in December and early<br />
January. Which resulted in especially cold ground temperatures.<br />
In April <strong>20</strong>02, MAGST at all sites reached their lowest<br />
levels in the last 5 years at least.<br />
A detailed comparison of the different sites reveals some<br />
peculiarities which can be attnbuted to local differences in interannual<br />
snow conditions.<br />
LOCAL SCALE<br />
Delaloye, R. et al.: Snow effects on recent shifts.<br />
CONCLUSIONS<br />
The analyzed data illustrate the interannual influence of snow<br />
conditions on the annual ground temperature. MAGST can<br />
vary sharply both in time and space without any change in<br />
MAAT.<br />
lnterannual variations in snow cover conditions at a regional<br />
scale can induce strong changes in the ground surfac<br />
temperature, which can be as high as I .5"C for<br />
MAGST calculated over an entire area. Locally, or in mountainous<br />
regions subject to more dramatic interannual shifts in<br />
weather conditions than in the western Swiss Alps, interannual<br />
changes in MAGST could be expected to be much<br />
higher.<br />
Interannual differences in predominant winds can result<br />
locally in completely different snow cover distribution which<br />
in turn has an influence the on spatial pattern of MAGST.<br />
The above conclusions indicate in particular that singleyear<br />
measurement of the MAGST in high mountain areas<br />
must be very carefully interpreted.<br />
The example illustrated on Figure 2 shows the effect of interannual<br />
changes in snow conditions on the MAGST at a very<br />
local scale.<br />
The four UTL-1 involved were located on a complex of<br />
push-moraines. In contrast with boththeprecedingand the<br />
following years, the winter of <strong>20</strong>00101 was characterized by<br />
predominantly southerly winds which drastically modified the<br />
snow distribution in the area. As a result, the spatial pattern d<br />
the MAGST was inversted during <strong>20</strong>01. The coldest places<br />
at the end of <strong>20</strong>00 became the warmest a few months later,<br />
before being the coldest again in <strong>20</strong>02. While the contrasted<br />
thermal evolution at Ri-19 and Ri-<strong>20</strong> can be easily explained<br />
by their location on opposite sides of a pass, the behaviour of<br />
Ri-21 and Ri-22 appears to be due to the presence of<br />
plurimetric-sized ridges and furrows at that place. Variations<br />
in the redistribution of snow by wind over a very short distance<br />
- a few metres - is seen as being responsible for this<br />
peculiar thermal behaviour.<br />
REFERENCE<br />
Vonder Muhll, D., Stucki, T. and Haeberli, W. 1998. Borehole<br />
Temperatures in Alpine Permafrost : A Ten-Year Series.<br />
<strong>Proceedings</strong> of the 7Ih International Conference on Permafrost,<br />
Yellowknife, Canada. 1089-1095.<br />
1 .0<br />
h<br />
0.0<br />
E<br />
2<br />
-1.0<br />
V'I -2.0<br />
<strong>20</strong>00 <strong>20</strong>01 <strong>20</strong>02<br />
Figure 2. Monthly running MAGST near the Six Rouge pass at the<br />
Ritord site. On the location map, the white area corresponds to<br />
loose sediments (mainly push-moraines). Outcropping bedrock is<br />
drawn in grey. The Six Rouge is the small summit west of Ri-19.<br />
24
Dangerous engineering-cryogenic processes at work in the urban<br />
territories of Russia’s permafrost zone<br />
E. A. Domnikova<br />
Moscow State University, Faculty of Geography, Moscow, Russia<br />
INTRODUCTION<br />
Vast territories in Russia are located in permafrost zones.<br />
Large natural resources are concentrated there and consequently<br />
many buildings and facilities have been constructed.<br />
Perennially frozen grounds are unstable with<br />
temperature fluctuations and readily change their condition<br />
from frozen to thawed and vice versa. These<br />
changes are followed by a dangerous intensification of<br />
geocryogenic processes with negative consequences for<br />
the geoecological situation.<br />
Cryogenic processes are geological processes which<br />
are related to the perennial and seasonal freezing of<br />
rocks (Kudjavtsev 1978).<br />
The following main groups of natural cryogenic processes,<br />
differing in their directions of influence and their<br />
specific manifestations in Far North landscapes, can be<br />
singled out: processes connected with warming effects<br />
on permafrost (thermokarst, thermoerosion); processes<br />
caused by cooling influences on the ground such as<br />
thawed-ground freezing or additional ice segregation in<br />
frozen ground (frost heave, cryogenic cracking); specific<br />
cryogenic processes (slope processes, the process of<br />
cryogenic weathering and the formation of frozen<br />
grounds under specific salt conditions).<br />
Natural cryogenic processes increase with industrial<br />
development in permafrost zones and often new engineering-cryogenic<br />
processes arise. In a lot of cases these<br />
engineering-cryogenic processes produce disastrously<br />
negative influences on the environment, buildings and<br />
structures.<br />
All engineering-cryogenic processes can be divided<br />
into two major types: those having a natural analog and<br />
their counterparts that have no complete natural analog.<br />
The first type includes activated natural cryogenic processes<br />
and also those caused by anthropogenic activity but<br />
also occurring under natural conditions. The second type<br />
embraces processes that are not found in nature.<br />
To classify the processes belonging to the first type we<br />
can use the groups cited for natural cryogenic processes<br />
but adjusted for technogenic impact. Given below is a<br />
classification of engineering-cryogenic processes that<br />
have no natural analog since these processes are most<br />
widespread in industrially developed territories of the permafrost<br />
zone.<br />
Technogenic heating of the ground is a process that<br />
increases the temperature of the perennially frozen rock.<br />
This process is intensified in industrially developed territories<br />
if the needs of local environment are violated by the<br />
maintenance of the facilities.<br />
Technogenic salting of the ground is the process<br />
whereby pollutants coming into the active layer alter the<br />
properties of the ground. Frequently, under such conditions<br />
of deep thawing within the boundaries of cities and<br />
settlements cryopegs arise - unfrozen strongly mineralized<br />
horizons with below zero temperatures. Technogenic<br />
salting results in the active destruction of the foundation<br />
material (Grebenets et a1.1997).<br />
TECHNOGENIC INUNDATION<br />
Most frequently, technogenic inundation is caused by<br />
leaks from engineering structures. This process weakens<br />
the ground structure and lowers its strength. It often involves<br />
ground ice which leads to thermokarst and thus<br />
deforming structures and foundations (Fig.1).
DESTRUCTION OF CONCRETE BY FROST<br />
Domnikova, EA: Dangerous enginee~ing-cryogenic processes at work<br />
The concrete of foundation pillars and supports is slowly<br />
destroyed under repeated freezing - thawing conditions of<br />
the moisture contained in the pores, microcrevices and<br />
caverns (Grebenets 1995). When the water contained in<br />
the cracks is converted into ice it expands and widens the<br />
crack that is undergoing the freezing. This major factor,<br />
producing negative effects on the foundation material, is<br />
increased as the depth of thawing increases due to<br />
elevated ground temperatures caused by technogenic<br />
sources of heating.<br />
It would seem advisable to classify not only the engineering-cryogenic<br />
processes but also the degree of their<br />
impact. This whole set of factors, leading to increasing<br />
engineering-cryogenic processes, can be conditionally divided<br />
into two groups: passive effects on the perennially<br />
frozen rocks and active effects on the perennially frozen<br />
rocks. Thus, all the engineering-cryogenic processes can<br />
be divided into two groups: engineering-cryogenic processes<br />
that only emerge under active effects on perennially<br />
frozen rocks and engineering-cryogenic processes that<br />
emerge even under passive effects on perennially frozen<br />
rocks.<br />
Various engineering-cryogenic processes occur with<br />
different intensities under the same conditions of technogenic<br />
effect. It would seem relevant to single out four different<br />
categories of intensity:<br />
I .Engineering-cryogenic processes with no marked<br />
effects;<br />
2.Engineering-cryogenic processes producing low intensity<br />
effects;<br />
3,Engineering-cryogenic producing high intensity effects;<br />
4.Engineering-cryogenic processes producing disastrous<br />
effects;<br />
Engineering-cryogenic processes can be controlled by<br />
stopping or neutralizing the processes of the technogenic<br />
impact, or by taking special engineering measures so that<br />
the geosystem reverts to its natural state, or by building<br />
new properties to the requirements of the geoecology and<br />
geocryology. Once this has become possible,<br />
engineering-cryogenic processes are considered to be<br />
controllable. If not, they are uncontrollable.<br />
Studies of the dynamics of dangerous engineeringcryogenic<br />
processes, during the industrial development of<br />
the area have revealed their great dependence on regional<br />
geocryologic peculiarities as well as on the character<br />
and intensity of the technogenic impacts applied.<br />
The map "The Intensification of Engineering-Cryogenic<br />
Processes on the Urban Territories of Russia's Permafrost<br />
Zone" (I:I<br />
5'000'000) has been created for the first time. It<br />
has been based on the "Geocryological Map of the USSR<br />
and the following parameters were taken into consideration:<br />
The character of the permafrost rocks in the area<br />
(solid. unsolid); whether it is flat or mountainous territory;<br />
the permafrost ground temperature and the level of zero<br />
annual amplitude; the population of an inhabited area; the<br />
26<br />
type of industrial development and the predominant engineering-cryogenic<br />
processes.<br />
The problems of stable developments in permafrost<br />
zones are very important now.<br />
REFERENCES<br />
Kudrjavtsev, V.A. 1978. General Geocryology. Moscow State University<br />
Publisheryouse (in Russian).<br />
Grebenets, V.I., Kerimov, A.G.-o. and Savtchenko, V.A. 1997. Stability<br />
of foundation in cryolitic zone under the condition of technogenic<br />
inundation and salting. Publications Committee of XIV<br />
ICSMFE (ed). Proc. XlVth Int. Conference on Soil Mechanics and<br />
Foundation Engineering, Hamburg, 6 - 12 September 1997: 113-<br />
114. Rotterdam: Balkerna.<br />
Grebenets V.I. 1995. Heat regime operation of frozen grounds on urban<br />
areas. New constructions of foundaments and methods of<br />
foundament preparation [Novyye konstruktsii fundamentov I metody<br />
podgotovki osnovaniy]. VNllOSP Proc., VOI. 95. Moscow: 66-<br />
79 (in Russian).
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Environmental maps of the Russian Arctic as a base for the segmentation<br />
of the coastline and the analyses of coastal dynamics<br />
D. S. Drozdov and Yu. V. Korostelev<br />
Earth Cryosphere Institute, SB RAS, Russia<br />
During the past years coastal dynamics in the Arctic Seas<br />
have been studied under different aspects by various scientific<br />
divisions and organizations. The main task of the<br />
project is to investigate fundamental scientific questions<br />
concerning coastal dynamics in the Eurasian Arctic by<br />
using Geographic Information System (GIS) technology -<br />
mapping, zoning and monitoring of the caostal exogenic<br />
processes, based on geosystem principles.<br />
Modern conditions and dynamics of the natural geosystem<br />
in northern regions with permafrost areas exist<br />
and constantly change, influenced by natural factors and<br />
anthropogenic activities. To study the influence of these<br />
factors, several GIS have been developed which can be<br />
applied to produce cartographical models of the main<br />
components of the geosystem. Digital environmental<br />
maps at global, regional and local scales are the most<br />
significant form of spatial realization of these models.<br />
Various prognostic, eco-geological and environmental<br />
maps were created and are being further developed at the<br />
present time.<br />
The GIS technology can be used to handle a geocryological<br />
data base correlated with particular polygons on<br />
the map and to create digital landscape, engineeringgeocryological,<br />
eco-geological, environmental, etc. maps.<br />
The GIS technology also supports monitoring needs, especially<br />
in the case of significant human activity. If a forecasting<br />
algorithm is accepted based on designed scripts of<br />
natural dynamics and economic activity, the GIS and<br />
monitoring system may represent not only modern data,<br />
but also prognostic information on environmental parameters.<br />
To analyze the dynamics of the Russian Arctic coastal<br />
zone a set of digital landscape and Quaternary geology<br />
maps with scales of 1:4 000 000 - 1 :2 500 000 for the hard<br />
copy are currently being developed. The first draft of these<br />
maps (Fig.1) was published within the framework of the<br />
CAVM project (Drozdov et al. <strong>20</strong>01, Walker et al. <strong>20</strong>02).<br />
According to the recommendations of the Arctic Coastal<br />
Dynamics (ACD) project (Rachold et al. <strong>20</strong>02) and the<br />
Russian World Ocean project, the mapped areas will be<br />
enlarged and more details on landscape and geological<br />
units will be provided, especially along the coastline. Several<br />
climatic zones from forest-tundra to central taiga will<br />
be included in the mapped area and new types of landscape<br />
and geological units with different lithology and<br />
permafrost extent appear on the map. The consequent re-<br />
processing of the source and derived maps, of satellite<br />
imagery and of the data base have to pass through several<br />
iterations.<br />
The basic information which is needed to create the<br />
environmental-geological GIS is collected in small-scale<br />
graphic documents presented in the literature and in archives.<br />
The most important information can be taken from<br />
maps published by I.S.Gudilin in 1980, G.S.Ganeshin in<br />
1976, J.Brown and others in 1997 and E.S.Melnikov and<br />
others in 1999. Additionally, the results of own research<br />
and the processing of remote sensing information were<br />
used.<br />
Figure 1. The landscape map of the Russian Arctic coastal zone: Climatic zonal, subzonal and altitudal-longtitudal landscape units.<br />
27
The major part of the information needed for the<br />
analyses of coastal dynamics can be extracted from the<br />
landscape map of the Russian Arctic coastal zone which<br />
provides a subdivision of landscape unit types. This<br />
principle of unit types is widely used for drawing up<br />
landscape, geocryological and environmental-geological<br />
maps of the northern territories of Russia. It can be applied<br />
for all scales, the only differences are the dimensions<br />
of the examined and displayed landscape and<br />
geological units. In the case under discussion, case we<br />
deal with the units of largest dimension. Each landscape<br />
- forming attribute affects on-shore exogenic processes<br />
including coastal dynamics. The main landscape forming<br />
controls are the following:<br />
Location of the area in the particular natural-climatic<br />
zone or subzone (e.g., arctic tundra, northern foresttundra,<br />
central taiga, etc.).<br />
Location of the territory in units of altitude zoning<br />
(i.e., plains, plateau, mountains).<br />
Genesis and main morphological attributes of relief<br />
(e.g., landscapes of marine plains and terraces, glacial<br />
landscapes, erosive-denudation landscapes of<br />
mountains and piedmonts, etc.).<br />
Characteristics of sediments: lithology of soils and<br />
petrography of rocks (e.g., clay, sand, debris material,<br />
karst-able and non-karstable bedrocks, etc.).<br />
Each of the listed attributes has to be mapped and<br />
the overlay of all these maps represents the final landscape<br />
map, which serves as the base for spatial characteristics<br />
of modern geological processes. Formally, all<br />
lines on the obtained landscape map can be estimated<br />
as the boundaries of common range. For convenience of<br />
analysis, however, it is useful to organize the boundaries<br />
in the order mentioned for the landscape-forming controls<br />
where the boundaries of natural-climatic zones are<br />
the most important and the boundaries of ground types<br />
are least important.<br />
The landscape and surface geology maps obtained<br />
serve as a graphical base to make segmentation of the<br />
Russian Arctic coastlines for description and analysis of<br />
coastal processes and for further assessment of coastal<br />
dynamics (Fig. 2).<br />
At the present time, segmentation of one coastal zone<br />
is practically completed. Each segment is to be supplied<br />
with a table of appropriative data according to the appropriate<br />
legend - the ACD classification template. This<br />
work is currently being performed by specialists from<br />
different institutes and organizations involved in the ACD<br />
project.<br />
ACKNOWLEDGEMENTS<br />
This study is supported by INTAS (grant 01-2332) and<br />
RFFl (grant 01-05-64<strong>25</strong>6).<br />
REFERENCES<br />
Circum-Arctic map of permafrost and ground ice conditions. 1997.<br />
Edited by J. Brown, O.J. Fenians, Jr., J.A. Heginbottom, E.S.<br />
Melnikov - IPA. US Geological Survey.<br />
Drozdov D.S., Ananjeva (Malkova) G.V. <strong>20</strong>01. Landscape Map of<br />
the Russian Arctic. I/ Forth Int. Circumpolar Arctic Veg. Map.<br />
Work: Moscow, April 10-12: 43-45.<br />
Melnikov E.S. (ed.) et alp, 1999: Landscape map of Russia permafrost<br />
at scale 1.4 000 000. Tyumen. Earth Cryosphere Institute<br />
S0 RAS (in Russian).<br />
Rachold, V., Brown, J., Solomon, S. <strong>20</strong>02. Arctic Coastal Dynamics<br />
- Report of an International Workshop, Potsdam (Germany) 26-<br />
30 November <strong>20</strong>01, Reports on Polar <strong>Research</strong> 413, 103 pp.<br />
Walker D.A. et al. <strong>20</strong>02. The circumpolar Arctic vegetation map. -<br />
Remote Sensing, V.23, #21: 4551-4570.<br />
Figure. 2. The segmentation of the Russian Arctic coastal zone based on landscape and surface geology maps.<br />
28
8th International Gonfcrence on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Use of BTS measurements and logistic regression to map permafrost<br />
probability in complex mountainous terrain, Wolf Creek, Yukon Territory,<br />
Canada<br />
M. Ednie and A.G. Lewkowicr<br />
Department of Geography, University of Ottawa, Canada<br />
Many thousands of square kilometers of north-western and <strong>20</strong>02. Miniature data loggers installed in fall <strong>20</strong>00<br />
Canada are underlain by mountain permafrost. At present, showed that stable minimum temperatures at the<br />
even the most up-to-date permafrost maps of this area are snow-ground interface were reached at both times of<br />
at scales which classify whole regions as sporadic or measurement, as required for the BTS methodology.<br />
widespread discontinuous permafrost, thereby ignoring 2) A map of potential incoming solar radiation for the<br />
the influence of elevation and exposure. In order to pro- basin was developed using a 30 m digital elevation<br />
vide baseline data for assessment of the effects of future model and the Solar Analyst extension (Fu and Rich<br />
climate change and to assist with mineral development, a 1999) in Arcview 3.2. The proportion of incoming rasimple<br />
methodology is needed to measure and identify diation modeled as direct beam was set using cloud<br />
permafrost distribution in this complex mountainous ter- cover data from Whitehorse.<br />
rain. The BTS (basal temperature of snow) method, widely 3) Measured BTS values were regressed against elevaemployed<br />
in the European Alps (e.g. Hoelzle et al. 1993) tion and the modeled potential incoming solar radiaand<br />
elsewhere, could be of great use in this context since tion and gave a statistically significant relationship<br />
it allows the development of a model of permafrost likeli- with an r*of 0.37. This value is comparable to those<br />
hood within a GIs. However, a thick, stable snow cover (at from European investigations (Gruber and Hoelzle<br />
least 80 cm deep) is required to insulate the ground from <strong>20</strong>01). The scatter in the graph of measured vs.<br />
atmospheric temperature variations in order to use BTS modeled BTS values (Fig. 1) likely derives from varitemperatures<br />
as indicators of permafrost. Studies of the ables that are not included in the modeling, such as<br />
thermal influence of the relatively thin and wind-blown albedo and vegetation cover.<br />
snow cover that is typical of northern Canada (e.g. Gran- 4) We determined the relationship between modeled<br />
berg 1973) have led to skepticism among some research- BTS and permafrost distribution, rather than employers<br />
as to the potential efficacy of the BTS method in this ing the categories used in the European Alps. In late<br />
region. Consequently, to our knowledge, this is the first <strong>July</strong>-August <strong>20</strong>02, holes were augered or pits were<br />
attempt to test the BTS method to map mountain perma- excavated at a total of <strong>20</strong>0 sites with a range of elefrost<br />
in North America.<br />
vations and slope orientations to verify the presence<br />
Our study was carried out in the Wolf Creek research or absence of frozen ground. Frozen ground enbasin,<br />
<strong>20</strong>-30 km south of Whitehorse in an area mapped countered within 2 m of the surface, or predicted from<br />
as part of the sporadic discontinuous permafrost zone. instantaneous temperature profile measurements to<br />
The basin has an area of almost <strong>20</strong>0 km2 at elevations of be present in this layer, was taken to be an indication<br />
750 to 2<strong>25</strong>0 m a.s.1. and a vegetation cover that changes of permafrost.<br />
from boreal forest to alpine tundra as elevations increase. 5) Logistic regression (e.9. Pereira and ltami 1991) was<br />
The annual air temperature at Whitehorse is -0.7OC used to evaluate the relation between the modeled<br />
(1971-<strong>20</strong>0) at 703 m a.s.1. and a value of -4OC was meas- BTS values and the permafrost distribution identified<br />
ured from <strong>20</strong>01-02 at an elevation of 1235 m within the in the excavations. A cluster of points from a confined<br />
basin.<br />
valley in the central part of the basin were significant<br />
The methods used to evaluate the efficacy of the BTS outliers and indicated that permafrost was present<br />
methodology to predict the probable distribution of perma- there at relatively low elevations. It is believed that<br />
frost in this complex mountainous terrain were as follows: these resulted from winter air temperature inversions<br />
which are common in the Wolf Creek basin. These<br />
1) A geo-referenced set of 396 BTS measurements points were excluded from the analysis. Traditional<br />
were collected within the basin in March-April <strong>20</strong>01 BTS results (e.g. Hoelzle et al. 1993) are given as<br />
29
-10 -8 -6 -4 -2 0<br />
Measured BTS (“C)<br />
Ednie, M. el aL: Use of BTS measurements and logistic regression ...<br />
qualitative values (permafrost improbable, permafrost between BTS values and permafrost probability (Fig. 2)<br />
possible, and permafrost probable) whereas logistic needs to be tested in other parts of the Yukon Territory.<br />
regression allows for the calculation of quantitative<br />
probabilistic values of permafrost distribution based<br />
on BTS (Figure 2). For Wolf Creek, there is a >90% REFERENCES<br />
probability of permafrost at modeled BTS values c-<br />
5OC, and -I .I‘C. Analyst: Arcview an extension for modeling at so<br />
scape scales. Accessible at:<br />
A preliminary probability permafrost map Of Wolf<br />
http://www.esri.com/librarvluserconf/~roc99/~roceed/~a~ers/pap8<br />
Creek basin was produced (not shown). From the 671~867.htrn<br />
analysis, about 40% of the basin is underlain by Per- Granberg, H.B. 1973: Indirect mapping of the snow cover for permamafrost.<br />
frost prediction and In <strong>Proceedings</strong> Scheffew<br />
Second International Conference on Permafrost, North American<br />
Contribution, Yakutsk, USSR. Washinaton, - DC: National Academy<br />
of Science Publication: 113-1<strong>20</strong>.<br />
Gruber, S. and Hoelzle, M. <strong>20</strong>01: Statistical modeling of mountain<br />
permafrost distribution: local calibration and incorporation of remotely<br />
sensed data. Permafrost and Periglacial Processes 12: 69-<br />
77.<br />
Hoelzle, M., Haeberli, W. and Keller, F. 1993: Application of BTS<br />
measurements for modeling mountain permafrost distribution. In<br />
<strong>Proceedings</strong> of the Sixth International Conference on Permafrost,<br />
Beijing. South China University of Technology, Beijing. Vol.1: 272-<br />
277.<br />
Pereira, J.M.C. and Itami, R.M. 1991: GIS-based habitat modeling<br />
using logistic multiple regression: A study of the Mt. Graham red<br />
squirrel. Photogrammetric Engineering & Remote Sensing 57(11):<br />
1475-1486.<br />
Figure I Measured BTS versus modeled BTS values.<br />
1 r--<br />
0.9 -<br />
e<br />
0.8 -<br />
0.7 -<br />
0.6<br />
n<br />
IC 0.5<br />
0<br />
-.-<br />
0.4 -<br />
n<br />
n<br />
n<br />
0.1 -<br />
0 4” ,<br />
-11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0<br />
BTS (“C)<br />
Figure 2: The probability of permafrost curve.<br />
We conclude that BTS modeling to predict permafrost<br />
is possible in mountainous parts of the Yukon Territory<br />
and that a quantitative probabilistic approach is the most<br />
logical given the variable terrain, vegetation and snow<br />
cover in this basin and elsewhere. Future research is<br />
needed to identify the topographic characteristics of<br />
mountain valleys that may experience air temperature inversions<br />
and to incorporate such characteristics into a<br />
permafrost prediction model. In addition, the relationship
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Modeling the influence of moisture variations on physical rock weathering<br />
processes in Antarctica: proposed investigations and preliminary results<br />
c. Elliott<br />
Geography Department, University of Canterburyl Christchurch, New Zealand<br />
The primary objective of the proposed research is to<br />
quantify the role of moisture in the weathering of rock in a<br />
cold environment such as Antarctica. It is being conducted<br />
to fulfill the requirements of a Ph.D. Although primarily focussed<br />
on physical weathering processes, some investigations<br />
of the potential role of chemical and biological<br />
weathering will also take place. Specifically, the research<br />
aims to develop a model to predict changes in weathering<br />
rate, for a particular rock type, along a temperature and<br />
moisture gradient on the Victoria Land Coast, Antarctica<br />
(Figure IA). One of the more challenging aspects of investigations<br />
into rock weathering is the complexity of factors<br />
involved and Antarctica is an ideal environment to undertake<br />
such research as its low moisture (at least in the<br />
area under consideration here) and extremely low temperatures<br />
enable some of these factors to be excluded or<br />
minimized.<br />
Despite the fact that the weathering of rock is a fundamental<br />
landscape process, the important role that moisture<br />
plays in weathering, although recognized, has not yet<br />
been quantified. For example, there has been no systematic<br />
study of how increasing quantities of moisture may influence<br />
the type of process operating. Nor, apart from the<br />
work of Prick (1997), have attempts to quantify the influence<br />
of changing moisture availability on weathering rates<br />
been made. With one or two notable exceptions (e.g.<br />
Matsuoka 1991), very few rock weathering studies have<br />
attempted to develop models that will predict weathering<br />
rates. A series of simulations will be conducted on rock<br />
samples from four locations in the field using an environmental<br />
cabinet and the effects of moisture isolated from<br />
other factors. The use of an environmental cabinet and<br />
speeded-up climatic cycles is a well-established technique<br />
for measuring weathering rates (e.g. Warke and Smith<br />
1998). However, in this study the climatic cycles to be<br />
used in the laboratory will be based on field observations<br />
of temperature and moisture regimes within the rock,<br />
rather than those using air temperature, for example. To<br />
achieve this, water baths will be used to provide the<br />
moisture levels necessary and the cabinet fitted with a<br />
lamp as an artificial 'sun' to enable the rock surface temperatures<br />
to be achieved.<br />
An empirically-based mathematical model will then be<br />
developed that will predict changes in weathering rate depending<br />
on changes in moisture in the rock. The weathering<br />
rate will be estimated by both measuring changes in<br />
the weight of the samples and using a non-destructive<br />
sonic technique to detect changes in strength (e.g. Fahey<br />
and Gowan 1979), after a specific number of cycles.<br />
Four locations along the Victoria Land Coast have<br />
been identified to undertake the fieldwork: Victoria Valley,<br />
Gneiss Point, Terra Nova Bay and Cape Hallett (Fig. IB).<br />
These locations enable both a coast-to-inland perspective<br />
(between Victoria Valley and Gneiss Point at appioximately<br />
77.5 OS), as well as a 5' latitudinal gradient (between<br />
Gneiss Point at 77.5 'S and Cape Hallett, at 72.5<br />
O S ) to be investigated. Hourly recordings of surface and<br />
subsurface (45 mm, 90 mm) rock temperature measurements<br />
from mid-October to midJanuary will be made using<br />
thermocouples and dataloggers. One-minute temperature<br />
measurements will also be taken during each<br />
visit (Le., in mid-October and mid-January) in order to determine<br />
whether any thermal gradients likely to induce insolation<br />
weathering have occurred. Year-round measurements<br />
will be attempted during the second field season in<br />
<strong>20</strong>03104.<br />
Surface moisture sensors will provide measurements<br />
of hourly surface moisture data, and relative humidity<br />
probes will enable subsurface moisture regimes to be determined<br />
(at 45 mm and 90 mm depths). The latter technique<br />
has not yet been used to measure subsurface<br />
moisture content in the field in rock weathering studies,<br />
but has been used successfully for measuring moisture<br />
content in concrete. These measurements are recorded<br />
manually and so can only be undertaken during the times<br />
of visits.<br />
These techniques represent a unique aspect of this research,<br />
as previous studies, such as those undertaken by<br />
Hall (1986), have measured average moisture content of<br />
rock samples left in the field rather than investigating levels<br />
of moisture at different depths in the rock. In addition,<br />
macro-climatological data (air temperature, relative humidity,<br />
wind speed and direction) provided through existing<br />
Automatic Weather Stations will be analyzed.<br />
31
~<br />
~~<br />
~ ~<br />
Table 1: Mean vapor densities at Gneiss Point October <strong>20</strong>02 and<br />
January <strong>20</strong>03 at 45 mm and 90 mm depth in the rock<br />
Depth of Vapor Density (gmlm2)<br />
Probe October January<br />
45 mm 1.4 5.1<br />
90 1.2 mm<br />
5.2<br />
Figure 2. Vapor densities for the Gneiss Point Site at 45 mm and 90<br />
mm depths in January <strong>20</strong>03 indicate that the 90 mm vapor densities<br />
are greater than the 45 mm vapor densities during the evening and<br />
early morning, but are less than the 45 mm ones during the day.<br />
REFERENCES<br />
Island<br />
Fahey, B. D. and Gowan, R.J. 1979. Application of the sonic test to<br />
experimental freeze-thaw studies in geomorphic research. Arctic<br />
and Alpine <strong>Research</strong> 11 (2): <strong>25</strong>3-260.<br />
Hall, K. 1986. Rock moisture content in the field and laboratory and its<br />
relationship to mechanical weathering studies. Earth Surface<br />
Processes and Landforms 11 : 131-142.<br />
Matsuoka, N. 1991. A model of the rate of frost shattering: application<br />
to field data from Japan, Svalbard and Antarctica. Permafrost and<br />
Periglacial Processes 2: 271-281,<br />
Prick, A. 1997. Critical degree of saturation as a threshold moisture<br />
level in frost weathering of limestones. Permafrost and Periglacial<br />
Processes 8: 91-99.<br />
Warke, P. A. and Smith, B.J. 1998. Effects of direct and indirect heating<br />
on the validity of rock weathering simulation studies and durability<br />
tests. Geomorphology 2: 347-357.<br />
Figure 1. (A) General location of field sites in relation to the Antarctic<br />
continent (6) The Victoria Land Coast indicating the four field sites at<br />
Victoria Valley, Gneiss Point, Terra Nova Bay and Cape Hallett. Not<br />
to scale.<br />
The first fieldwork season has just been completed at<br />
Gneiss Point and Victoria Valley and rock temperature<br />
and moisture data successfully captured. Initial results<br />
give a clear indication of the different temperature regimes<br />
between the October and January recordings, as well as<br />
the changes in moisture levels at the different depths<br />
within the rock. Mean moisture levels between October<br />
<strong>20</strong>02 and January <strong>20</strong>03, although similar between depths<br />
for each period, show significant differences between periods<br />
(Tablel) and during the course of the day (Figure 2).<br />
32
8th International Conference an Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available l ~ ~<br />
The lower discontinuous permafrost boundary in the Argentera Massif<br />
(Maritime Alps, Italy): insight from rock glacier geo-electrical soundings<br />
D. Fabre, *A. Ribolini, P. R. Federici and M. Pappalardo<br />
Laboratoire lnterdisciplinaire de Recherche lmpliquant la Geologie et la Mecanique (LIRIGM), Universite Joseph Fourier, France<br />
* Dipartimento di Scienze della Terra, University of Pisa, Pisa, Italy<br />
The rock glaciers of the Argentera Massif are located in<br />
the southernmost part of the European Alps, only 50 km<br />
from the Mediterranean Sea. Glacial and periglacial reconstructions<br />
in this Alpine region indicate out a rock<br />
glacier formation mainly during the Younger Dryas and<br />
Subboreal cold phases (Finsinger and Ribolini <strong>20</strong>01 and<br />
references therein). In this marginal area of the Alps,<br />
transitional climate between Mediterranean and more<br />
continental conditions made uncertain the evaluation of<br />
rock glacier activity, that in some cases seems controlled<br />
by local microtopography and meteorology<br />
(Federici et al. <strong>20</strong>00). Recently, Surface Ground Temperature<br />
(SGT) of some rock glaciers (Ribolini <strong>20</strong>01),<br />
allowed to site the Lower Discontinuous Permafrost<br />
Boundary (LDPB) at about 2,600 m a.s.l., which is<br />
slightly lower than the one proposed for the southern<br />
French Alps in the opposite flank of the mountain chain<br />
(Evin and Fabre 1980).<br />
To improve this evaluation, during the years <strong>20</strong>01-02<br />
geo-electrical soundings were conducted on four rock<br />
glaciers (namely: Bassa Margot, Agnel, Portette and<br />
Prefouns), selected at mean elevations ranging between<br />
2,450 m and 2,700 m, thus proximal to the inferred<br />
LDPB. All the analyzed rock glaciers belong to the<br />
Gesso Basin in the southeastern Argentera Massif.<br />
The Schlumberger method was adopted (with AB12<br />
max ranging from <strong>20</strong> to 70 m), using an apparatus<br />
(MAATEL BMI) working at DC <strong>20</strong>00 V of max power.<br />
This device was calibrated on the Murtel rock glacier in<br />
<strong>Switzerland</strong>, where a mechanical sounding through the<br />
permafrost is available (Fabre and Evin 1990). Previous<br />
soundings in the southern French Alps (Assier et al.<br />
1996 and references therein) evidenced a typical active<br />
rock glacier vertical profile of resistivity: a) surface layer<br />
without permafrost (5,000-<strong>20</strong>,000 Dm), b) ice-saturated<br />
sediments (permafrost) with resistivity ranging from tens<br />
of thousand up to million of Wm, and c) bedrock or not<br />
frozen basal sediment (1000-10,000 nm).<br />
Two longitudinal profiles in the upper-middle part<br />
(BMI, BM2) and a transversal one (BM3) in the lower<br />
part were performed on the Bassa Margot rock glacier.<br />
The results (Fig. 1) indicate presence of permafrost with<br />
medium-high ice content layers in the upper-middle part,<br />
while in the lower one unfrozen sediments were found.<br />
The Portette rock glacier also shows high resistivities.<br />
In the upper part transversal profile (POI), data are<br />
consistent with a 5 m thick permafrost level with a medium<br />
ice content (110,000 Om), overlaying a <strong>20</strong> m thick<br />
of permafrost with an higher ice content (-1 Mnm).<br />
Similar resistivities in the vertical distribution results from<br />
the transversal profile in the lower part (P02) (Fig. 1).<br />
The three transversal profiles (PRI, PR2, PR3) performed<br />
on the Prefouns rock glacier show the presence<br />
of permafrost, with thick layers characterized by low to<br />
medium ice contents. Here, the results of the soundings<br />
are consistent with a permafrost thickness of <strong>20</strong> m<br />
overlaid by a few metres of sediments with little ice<br />
content (Fig. I).<br />
Agnel rock glacier was investigated by means of only<br />
one transversal profile (AG), because of its limited areal<br />
extension. The resistivity distribution found (Fig. 1) suggests<br />
a thin layer of permafrost (I to 7 m depth), with<br />
low ice content (70,000 nm) probably not in equilibrium,<br />
characterized by a temperature near the melting point (T<br />
> -1 "C)<br />
Considering that the rock glaciers studied could be<br />
considered representative of all the rock glaciers that are<br />
present in the Argentera, a speculation about activity<br />
elevation can be proposed. Sectors of rock glacier developed<br />
above 2,600 m are affected by permafrost, up to<br />
<strong>20</strong> m thick and with medium-high ice content (<strong>20</strong> to 40<br />
%). Rock glaciers at elevations ranging between 2,450<br />
and 2,600 m are affected by thinlrelict permafrost layers.<br />
Below 2,450 m it is unlikely to find permafrost, except for<br />
very isolated, sheltered sites.<br />
On the basis of these results, the Argentera Massif<br />
can be considered the southernmost discontinuous permafrost<br />
environment of the Alps, whose lower boundary<br />
corresponds to about 2,600 m a.s.1.<br />
33
Fabre, Q. et at, : Permafrost boundary, Argentera Massif, rack glacier, geo-electrical soundings.<br />
I<br />
REFERENCES<br />
Fabre, D. and Evin, M. 1990. Prospection electrique des milieux a<br />
tres forte resistivitk le cas du pergklisol alpin. Proceeding of<br />
Sixth International Congress of the International Association of<br />
Engineering Geology, Pays-Bas: 8.<br />
Assier A,, Fabre, D. and Evin, M. 1996. Prospection Blectrique sur<br />
les glaciers rocheux du cirque de Ste-Anne. Permafrost and<br />
Periglacial Processes 7: 53-68<br />
Finsinger, W. and Ribolini, A. <strong>20</strong>01. Late glacial to Holocene deglaciation<br />
of the Colle del Vei del Bouc-Colle del Sabbione Area<br />
(Argentera Massif, Maritime Alps, Italy-France). Geografia Fisica<br />
e Dinamica Quaternaria 24: 141-156.<br />
Federici, P.R., Pappalardo, M. and Ribolini, A. <strong>20</strong>00. On the ELA<br />
and lower limit of discontinuous permafrost boundary in the<br />
Maritime Alps (Italian side). Atti Accademia delle Scienze di Torino<br />
134: 23-29.<br />
Ribolini, A. <strong>20</strong>01. Active and fossil rock glaciers in the Argentera<br />
Massif (Maritime Alps): Surface ground temperatures and paleoclimatic<br />
significance. Zeitschrift fur Gletscherkunde und Glazialgeologie<br />
37: 1<strong>25</strong>-140.<br />
34
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available information<br />
Using catchment soils to establish the provenance of high arctic pond<br />
sediments: Truelove Lowland, Devon Island, Canada<br />
LA. Fishback and *R.H. King<br />
Scientific Coordinator, Churchill Northern Studies Centre, Canada<br />
*Deparfment of Geography, The University of Western Ontario, London, ON, Canada N6A 5C2<br />
INTRODUCTION<br />
High arctic ecosystems have been identified as critically<br />
sensitive areas in terms of global changes because of the<br />
strong negative feedback mechanisms that create sensitivity<br />
to the effects of greenhouse warming. Lakes and<br />
ponds in these regions have been identified as “sensitive<br />
bellweath-ers” of these changes (Douglas et al. <strong>20</strong>00).<br />
Ponds are defined as water bodies that freeze to the<br />
bottom each winter whereas lakes maintain an unfrozen<br />
water column during the winter months.<br />
The preservation of the paleo-environmental record in<br />
arc-tic pond sediments has been found to be more complete<br />
than anticipated (Douglas et ai. <strong>20</strong>00). However,<br />
such paleo-environmental reconstructions tend to be far<br />
more effective in determining past stages of the water<br />
body than providing indi-cations of environmental changes<br />
in the surrounding catch-ment areas.<br />
This research attempts to address the catchment<br />
condi-tions with respect to pond sediment records by<br />
gaining a bet-ter understanding of the physical, chemical<br />
and mineralogical signal of catchment processes and<br />
identifying the presence of this signal in the adjacent high<br />
Arctic pond sediments.<br />
STUDY AREA<br />
Truelove Lowland (75040’N, 84035’W) is one of a series<br />
of coastal lowlands with high biological diversity located<br />
on northeastern Devon Island, Nunavut, Canada. As a result<br />
of isostatic emergence following the last glaciation, a<br />
series of raised beaches was formed (King 1991). These<br />
features are topographic barriers that have trapped water<br />
resulting in the formation of approximately 530 ponds and<br />
five lakes (ap-proximately <strong>20</strong>% of the Lowland area) (King<br />
1991). Pooh Pond in the central Lowland and the adjacent<br />
raised beach were selected for this study.<br />
METHODS<br />
The sediments of Pooh Pond, a large, 1.2 m deep,<br />
freshwa-ter pond were characterized using four short<br />
sediment cores (e 0.40 m) taken from a variety of water<br />
depths in a transect from shore to deepest area of the<br />
pond. Sediments were cored using a modified corer (id =<br />
51 mm) and fractioned at 50 mm intervals. Fractions were<br />
analyzed using chemical fractionations and selective extractions<br />
to characterize the sediments and differentiate<br />
between materials from the catch-ment and the water<br />
body, as well as establishing indicators of specific catchment<br />
processes.<br />
Major soil inspection pits were excavated in a catena<br />
comprising four landscape units within the catchment. Soil<br />
ho-rizons from each pit were sampled in duplicate in the<br />
field and processed with the pond sediment cores. A data<br />
set was compiled of similar biogeochemical variables for<br />
pond sedi-ment and soils and tested for normalcy. A series<br />
of multivari-ate statistics were used to investigate the<br />
provenance of the pond sediments.<br />
RESULTS AND DISCUSSION<br />
A 14C pond sediment basal date of 5980 f 40 yrs BP<br />
(Beta-162840) from 0.32 m depth was comparable to a<br />
predicted emergence date for Pooh Pond, indicating a<br />
freshwater origin of pond sediments collected. The lack of<br />
biogeochemical stratigraphy in these pond sediment cores<br />
was most likely due to cryoturbation or wave-induced turbation<br />
of the sedi-ments that act to homogenize the unfrozen<br />
sediments. An ANOVA of the physical and chemical<br />
characteristics of the four cores demonstrated a significant<br />
difference between the four cores. These differences indicate<br />
the importance of local sedimentary environments in<br />
the pond and the difficulty of single core characterization<br />
of pond sediments.<br />
The four soil types investigated on the raised beach<br />
are formed on poorly stratified gravelly carbonate-rich<br />
beach de-posits (Lev and King 1999). Soils on the raised<br />
beach crest were poorly developed with desert pavement<br />
at the surface and a thin A horizon overlying a coarse parent<br />
material with carbonate pendants and particle caps.<br />
Brunification dominates the soils in the upper and lower<br />
foreslope as indicated by the Na-pyrophosphate extractable<br />
Fe in the soils in Table 1. These soils are more developed<br />
with high organic carbon contents, a thin Bm horizon<br />
and an overall finer texture. Contemporary<br />
translocation of Fe and AI in the lower foreslope soil soh-<br />
35
tions was indicative of incipient podzoliza-tion. Soils at the<br />
slope base are waterlogged with accumu-lated organic<br />
matter over a gleyed silty parent material with high concentrations<br />
of Mn and Fe.<br />
Table 1: Extractable Fe as an indicator of brunification in soil horizons<br />
in the raised beach soils.<br />
Soil Twe 1 Soil Horizons<br />
1 Feo 2<br />
(Landscape Unit)<br />
ReWosolic Static Cryosol Ahk<br />
(% wt)<br />
0.050<br />
(Raised Beach Crest) Cca 0.0<strong>25</strong><br />
C k2 0.0<strong>20</strong><br />
Brunisolic Eutric Static Crvosol 0.080 Ah<br />
(Upper Foreslope) 0.050 Bmk<br />
Ckl 0.035<br />
C k2 0.0<strong>25</strong><br />
Brunisolic Eutric Turbic AhY Cryosol 0.110<br />
Fishback, LA. et 81.: Using catchment soils to establish the provenanca..<br />
Agri-Food Canada Publication 1646 (Revised).<br />
Douglas, M.S.V., Srnol, J.P. and Blake, W.Jr. <strong>20</strong>00. Summary of paleolimnological<br />
investigation of high arctic ponds at Cape<br />
Herschel, east-central Ellesmere Island, Nunavut. In M. Garneau<br />
and B. Alt (eds.), Environmental Response to Climate Change in<br />
(Lower Foreslope) BmY 0.1 15<br />
Ckyl 0.0<strong>25</strong><br />
Cky2 0.030<br />
Gleysolic Turbic OfY Cryosol 0.210<br />
(Meadow)<br />
W Y 0.050<br />
1 as determined by the Canadian Soil Classification System, 1998.<br />
2 Na-pyrophosphate extractable Fe to represent organically chelated<br />
Fe in the B horizon in % of soil weight.<br />
Cluster analysis was used to classify the raised beach<br />
soil horizons into four groups to create a reference set for<br />
comparison with the pond sediments. The ‘unknown’ pond<br />
sediments were classified into the soil groups using a discriminant<br />
function analysis (DFA). The inorganic pond<br />
sedi-ments were classified predominantly with soil horizons<br />
showing some weathering. This DFA also identified<br />
the most discriminating variables in classifying the pond<br />
sediment with the soils as total organic carbon, organically<br />
chelated Fe, AI and Mn and ‘plant available’ elements<br />
(Mehlich Ill extract).<br />
A second cluster analysis of the entire soil and pond<br />
sediment data set resulted in separate groups of pond<br />
sedi-ments and soil horizons with the exception of one<br />
group. This group contained soil horizons from the toe of<br />
the slope and from the pond sediment core closest to the<br />
shore highlighting the localized connection of soils and<br />
pond sediments.<br />
chemistry of sediment alone. Pond sediments are a result<br />
of a number of genetic processes operating together to<br />
produce a unique material. Modification of the catchmentderived<br />
material dur-ing transport and deposition obscures<br />
the catchment signal and therefore influences the interpretation<br />
of the sediment record.<br />
REFERENCES<br />
Agriculture and Agri-Food Canada, Soil Classification Working Group.<br />
1998. The Canadian Soil Classification System. Agri-culture and<br />
the Canadian High Arctic. Geological Survey of Canada Bulletin<br />
529.<br />
King, R.H. 1991. Paleolimnology of a polar oasis, Truelove Low-land,<br />
Devon Island, NWT, Canada. Hydrobiologia 214: 317-3<strong>25</strong>.<br />
Lev, A. and King, R.H. 1999. Spatial variation and soil development<br />
in a high arctic soil landscape: Truelove Lowland, Devon ls-land,<br />
Nunavut, Canada. Permafrost and Periglacial Processes. 9: 35-<br />
46.<br />
CONCLUSIONS<br />
Soil profile morphologies reflected a variety of pedogenic<br />
and cryogenic processes within a soil catena across a<br />
raised beach in Truelove Lowland. These processes included<br />
brunification and incipient podzolization.Sediments<br />
from the adjacent Pooh Pond were heterogeneous and<br />
the characteri-zation of these sediments based on single<br />
cores was difficult. Biogeochemical analyses indicate that<br />
Pooh pond sediments are predominantly catchmentderived<br />
in origin.<br />
This research indicates that while these pond sediments<br />
are derived mostly from catchment soil material, it<br />
is difficult to establish provenance based on the biogeo-<br />
36
~ I<br />
~<br />
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Atmospheric and geothermal conditions during thermal contraction of ice<br />
wedges, Bylot Island, eastern Canadian Arctic archipelago<br />
D. Fortier and M. Allard<br />
Centre d'ktudes nordiques, Universite Lava/, Quebec, EIK 7P4<br />
We studied thermal contraction cracking of ice-wedge<br />
polygons located on a terrace bordering a glacio-fluvial<br />
outwash on Bylot Island (730 09'38"N, 800 0043,21 W).<br />
Previous studies showed that the polygons of the study site<br />
are composed of a syngenetic ice-rich accumulation of insitu<br />
grown peat and wind-blown silts (Fortier and Allard<br />
<strong>20</strong>01).<br />
Air and ground temperature conditions preceding the<br />
rupture of electric cables buried in the active layer over ice<br />
wedges were analyzed for the period 1997 to <strong>20</strong>02. Frost<br />
cracking (breaking of electric cables 5 mm 0.d.) occurred<br />
from November to March with the largest number of breaks<br />
(45%) taking place in January, during the polar night. Due<br />
to the absence of solar radiation, air temperature variations<br />
are closely linked to the thermal properties of the air<br />
masses. In the Bylot Island region and in northeastem<br />
Baffin Island, the atmospheric depressions in the winter<br />
come mainly from the south (Maxwell 1981). Analysis d<br />
wind data recorded over the study site showed that frost<br />
cracking of ice wedges occurred mainly following the<br />
incursion and persistence of eastward and southward cold<br />
weather accompanied by winds of varying strength from<br />
the Byam Martin mountains and Navy Board Inlet.<br />
The rate of geothemal contraction in the polygons is<br />
related, in a non-linear way, to the length, range and speed<br />
of the cooling of the air temperature, the thickness of the<br />
snow cover, the sediment types and the ice content of the<br />
ground (MacKay 1993, MacKay and Burn <strong>20</strong>02). Crack<br />
openings occur when contraction deformation by creep in<br />
ice wedges is not fast enough to release the mechanical<br />
tensions created by the thermal contraction of the terrain.<br />
From 1997 to <strong>20</strong>02, the mean daily air temperature on<br />
days when cracking occurred was -36 "C, with air temperatures<br />
ranging from -27 to -40 "C. Frost cracking of the<br />
ground (electric cable ruptures) occurred a few hours to a<br />
few days after a drop in the mean hourly air temperature in<br />
the order of 7.9 "C over an average of eighteen hours at a<br />
mean air cooling rate of -0.45 "C/h. Maximum cooling mte<br />
was -0,9 "Clh when air temperature quickly dropped by<br />
18.6 "C in less than 24 hours (12-<strong>20</strong>00). Fifteen of the<br />
nineteen cracking events occurred when the mean daily air<br />
temperature was stationary or rising slightly since the pre-<br />
ceding day. This is due to the penetration time of the cold<br />
wave into the snow cover and the frozen ground. Table 1<br />
shows air temperature cooling rates during the days before<br />
selected frost cracking events, for the 1997 to <strong>20</strong>02 period.<br />
Table 2 shows mean, maximum and minimum ground<br />
temperature occurring during frost cracking days. During<br />
cracking episodes, the mean temperature at the ground<br />
surface (-2 cm) was -22.9 "C(I<br />
997-<strong>20</strong>02) and ranged from<br />
-14.7 "C to -29 "C. Geothermal data obtained from a<br />
thermistor cable, installed in permafrost in 1999, reveal that<br />
mean daily ground temperature at the top of permafrost (40<br />
cm under the ground surface) during frost cracking events<br />
(I 3) was -18.6 "C (<strong>20</strong>00-<strong>20</strong>02) and varied from -23.8 "C b<br />
-12.8 "C.<br />
1 YEAR I DAY / MONTH I ATCR ("C I h)<br />
1997<br />
1997<br />
1997<br />
21 - 22/02 (24)<br />
13 14/03 (15)<br />
14 - 15/03<br />
-0,61(-8,5"C/14h)<br />
-0,44 (-6,2"C/14h)<br />
-0,43 (-6,9"C/l6h)<br />
I ISM I IO- iil0l I -0.55 (-9.4"C/l7hl I<br />
I999<br />
I999<br />
9<br />
<strong>20</strong>00<br />
2OOO<br />
<strong>20</strong>00<br />
22/01 (26)<br />
23 - 24/02 (26,27)<br />
24 - <strong>25</strong>/02 (26,27)<br />
21/01 (23)<br />
23/01<br />
(13)<br />
-0,21 (4,9"C/23h)<br />
-0,36 (-9,3"C/26h)<br />
-0,58 (-7,O0C/I2h)<br />
-0,53 (-5,8"C/l I h)<br />
-0,69 (-6,2"C/09h)<br />
-0'33 (-7,3"C/22h) 12/02<br />
<strong>20</strong>01 (5)<br />
-0,51 (-7,2"C/14h) 04/02<br />
<strong>20</strong>01 26 - 27111 (28)<br />
-0,36 (12,4"C /34h)<br />
<strong>20</strong>01 27 28/11<br />
-0,26 ( SOC I 19h)<br />
37
Fortier, D. et al. : Atmospheric, conditions, thermal contraction, ice wedges, Bylot Island ...<br />
Table 1. ATCR : air temperature cooling rate preceding selected<br />
frost cracking events of the 1997 to <strong>20</strong>02 period. Frost cracking<br />
days in bold.<br />
Lachenbruch, A.H. 1962. Mechanics of thermal contraction cracks<br />
and ice wedge polygons in permafrost. Geological Society of<br />
America, New-York: 69.<br />
MacKay, J.R. 1993. Air temperature, snow cover, creep of frozen<br />
ground, and the time of ice-wedge cracking, western Arctic<br />
coast. Canadian Journal of Earth Sciences 30: 17<strong>20</strong>-1729.<br />
MacKay, J.R. and C.R. Burn. <strong>20</strong>02. The first <strong>20</strong> years (1978-1 979<br />
to 1998-1999) of ice-wedge growth at the lllisarvik<br />
experimental drained lake site, western Arctic coast, Canada.<br />
Canadian Journal of Earth Sciences 39: 95-111.<br />
Table 2. Mean, minimum and maximum daily ground<br />
temperatures ("C) at given permafrost depths, during frost<br />
cracking days (<strong>20</strong>00-<strong>20</strong>02).<br />
Cracking episodes that occurred from <strong>20</strong>00 to <strong>20</strong>02 had a<br />
mean thermal gradient in the active layer (0-40 cm) of 10.9<br />
"Clm that varied from 3.9 to 18.8 "Clrn. Deeper down, the<br />
mean thermal gradient was 3.34 "Clm (1.1 to 5.5 "Clm)<br />
between 40 and100cm,3.12 "Clm (-2.1 to 4.3 "Clm)<br />
between I and 2 m and 1.64 "Clm (0.7 to 3.<strong>25</strong> "Clm)<br />
between 2 and 3 m. According to our analysis of the mean<br />
ground cooling rate occurring at different depths before frost<br />
cracks openedin<strong>20</strong>00-<strong>20</strong>02,the maximum cooling zone<br />
was primarily located in the active layer, most of the time<br />
being in the upper part close to the surface (table 3). Between<br />
0 and <strong>20</strong> cm, the mean daily groundcooling rate<br />
was -0.<strong>25</strong> "Clday and varied from -0,l to -1,l"Clday.<br />
Since the tensile strength of the ice-rich peaty loess is<br />
greater than the tensile strength of wedge ice, frost cracks<br />
probably initiated at or near the top of permafrost, Le., the<br />
top of the wedges. (Lachenbruch 1962).<br />
DEPTH MDGCR<br />
Surface -0.<strong>25</strong><br />
-0.21<br />
Table 3. Mean daily ground cooling rate (MDGCCR; "Clday) at<br />
given permafrost depths during frost cracking days (<strong>20</strong>00-<strong>20</strong>02).<br />
REFERENCE<br />
Fortier, D. and Allard, M. <strong>20</strong>01. Cryostratigraphy and chronology<br />
of syngenetic ice wedge polygons, Bylot Island, Canadian<br />
Arctic archipelago. <strong>Proceedings</strong> 3Ist international Arctic<br />
workshop, University of Massachussetts: 31-32,
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Results of a first study on tree growth and soil characteristics at a cold site<br />
located below the limit of discontinuous alpine permafrost<br />
M. Freppaz, L. Celi, *V. Stockli and M. Phillips<br />
Swiss federal Institute for Snow and Avalanche <strong>Research</strong>, Davos Doe <strong>Switzerland</strong> and Laboratory <strong>Research</strong> Center on Snow and<br />
Alpine Soils, Italy<br />
“Swiss Federal lnsfjfufe for Snow and Avalanche <strong>Research</strong>, Davos Do$ <strong>Switzerland</strong><br />
INTRODUCTION<br />
Scree slopes with particularly cold ground temperatures at<br />
low elevations have been known for centuries in <strong>Switzerland</strong>.<br />
They were termed ‘wind holes’, ‘ice holes’ or<br />
‘weather holes’ by local farmers who recognized the potential<br />
of such locations for cooling milk and butter during<br />
the hot summer months. The sites are generally steep<br />
scree slopes located at the foot of high cliffs.<br />
Studies effected over the past 50 years have demonstrated<br />
that the ground is not frozen as a result of a cold<br />
microclimate at these locations, but that it occurs due to a<br />
particular underground air circulation phenomenon and<br />
due to the presence of an insulating vegetation cover<br />
(Delaloye and Reynard <strong>20</strong>01).<br />
A particularly interesting and important characteristic is<br />
that despite their occurrence in montane areas, these<br />
sites have subalpine and alpine vegetation, and that some<br />
plants, in particular spruce (Picea abies L. Karst) display<br />
signs of severely limited growth. The soils generally reveal<br />
a high organic matter content, with the formation of a humus<br />
mor type at the surface.<br />
In this preliminary study we analyzed the main properties<br />
of these cold soils and of tree growth in order to<br />
elaborate the first hypotheses about the contribution of soil<br />
characteristics to the reduced plant growth.<br />
METHODS<br />
The Creux du Van in the Swiss Jura, at 1<strong>20</strong>0 m a.s.l.,<br />
where the patches of dwarf trees are especially well developed,<br />
was selected as a test site. To compare soil<br />
characteristics and tree growth in areas with different evidence<br />
of permafrost occurrence the site was divided into<br />
two strata, corresponding to the prevailing tree growth<br />
form: I) stunted forest, and II) normal spruce forest.<br />
In each of the two strata a spot was selected where a<br />
soil profile was dug and tree cores were sampled. The<br />
distance between the two sampling points, situated at the<br />
same elevation and aspect, was about 50 m.<br />
Around each soil profile in every stratum, three trees of<br />
representative sizes were selected. The height of the trees<br />
was estimated and the tree stems were cored perpendicular<br />
to their axes at heights of 30 cm above the ground.<br />
In the laboratory, the wood cores were processed with the<br />
usual dendrochronological methods (Schweingruber<br />
1988).<br />
To record soil temperature, thermistors combined with<br />
dataloggers (UTL-1) were buried in pairs in the two soil<br />
profiles at a depth of about 30 cm (Oa horizon). The loggers<br />
recorded the temperature throughout the period November<br />
<strong>20</strong>01 to November <strong>20</strong>02.<br />
The soil pH was determined in a 1 :<strong>20</strong> soil-water suspension.<br />
Total soil carbon and nitrogen were measured<br />
using a THERMOQUEST NC <strong>20</strong>05 combustion analyzer.<br />
The microbial biomass C and N were determined by the<br />
fumigation technique. Humic acids (HA) and fulvic acids<br />
(FA) were extracted from the soil and the elemental C and<br />
N content determined. The N availability was determined<br />
by anaerobic incubation of soil samples under water in<br />
closed containers at 40°C for 7 days.<br />
RESULTS AND DISCUSSION<br />
The analysis of tree ring patterns revealed that the smallest<br />
trees of stratum I exhibit ages around 90 years, despite<br />
their small size (mean height = 3 m). Trees in stratum<br />
II are distinctly larger (mean height = <strong>25</strong> m), older<br />
(124 years) and have much more rapid growth rates.<br />
Soil profiles typically exhibit an upper soil horizon (Oi)<br />
of a fibric humus type, an intermediate horizon (Oe) of a<br />
hemic humus type and a lower horizon comprising a<br />
sapric humus type (Oa), which filled the interstices between<br />
blocks of limestone.<br />
According to the thickness of the active layer, estimated<br />
at 2 m under the dwarf trees (Delaloye and<br />
Reynard <strong>20</strong>01), the soil was classified as gelic Histosol<br />
(stratum I) (ISSS Working Group RB. 1998). The depth of<br />
the active layer under the normal forest was estimated<br />
39
greater and the soil was classified as sapric Histosol<br />
(stratum 11).<br />
In both strata the humus type in the Oi and Oe horizons<br />
comprises mainly Sphagnum residues, which actively<br />
acidify the soil upper horizons and hence are responsible<br />
for the low pH (4.2 - 4.3). The pH in the Oa<br />
horizon of all three strata is higher than in the upper horizon,<br />
reaching values of 7.3 to 7.4.<br />
Mean soil temperature (Oa horizon) between November<br />
<strong>20</strong>01 and November <strong>20</strong>02 (Fig. I) was higher in stratum<br />
II (+3.43 "C) than in stratum I (t2.63 "C) as expected<br />
by the depth of the active layer.<br />
To test these hypotheses, and to better understand soil<br />
chemical status and plant reaction, further investigations<br />
are underway.<br />
REFERENCES<br />
Dalias, P., Anderson, J.M., Bother, P. and Couteaux, M.M. <strong>20</strong>01.<br />
Long-term effects of temperature on carbon mineralisation processes.<br />
Soil Biology & Biochemistry 33: 1049-1057.<br />
Delaloye, R. and Reynard, E. <strong>20</strong>01. Les eboulis geles du Creux du<br />
Van (Chaine du Jura, Suisse). Environnements pbriglaciaires (Association<br />
Franqaise du Periglaciaire) 8: 118-129.<br />
ISSS Working Group RB. 1998. World Reference Base for Soil Resources.<br />
International Society of Soil Science (ISSS), International<br />
Soil Reference and Information Centre (ISRIC) and Food and Agriculture<br />
Organization of the United Nations (FAO). Report No.84,<br />
Rome, 91.<br />
Myrold, D. 1987. Relationship between microbial biomass nitrogen<br />
and a nitrogen availability index. Soil Science Society of America<br />
Journal 51: 1047-1049.<br />
Schweingruber, F.H. 1988. Tree Rings: Basics and Application of<br />
Dendrochronology. D. Reidel Publishing, Dordrecht.<br />
-4 .J<br />
<strong>20</strong>0 1 <strong>20</strong>02<br />
Figure 1. Soil temperature recorded in the two soils (Oa horizon)<br />
(stratum I grey; stratum II black) from November <strong>20</strong>01 to November<br />
<strong>20</strong>02. Average values of the 2 loggers in each stratum.<br />
Soil organic matter content is high in both strata, due to<br />
the low soil temperature which may have contributed to<br />
reducing the C decomposition rate. The Corg and Norg<br />
concentrations of the Oa horizon were respectively equal<br />
to 382 g C kg-I and 15.9 g N kg-I in stratum I and 428 g<br />
C kg-I and 12.9 g N kg-I in stratum II. In the Oa horizon<br />
of stratum I the C and N content of the humic fraction<br />
(70.3 g C kg-I; 7.53 g N kg-I) was lower than in stratum II<br />
(103.6 g C kg-I; 8.08 g N kg-I), revealing how temperature<br />
might affect the biochemical pathways and decomposition<br />
in addition to their metabolic rates (Dalias et al.<br />
<strong>20</strong>01).<br />
The microbial C and N pools are significantly lower in<br />
stratum I (1.1 g C kg-1; 0.12 g N kg-I) than in stratum II<br />
(17.5 g C kg-I ; 2.50 g N kg-I).<br />
The N availability assessed by incubation under anaerobic<br />
waterlogged conditions was equal to 29.2 mg N<br />
kg-I in stratum II and 14.8 mg N kg-I in stratum I. The<br />
amount of mineralizable N obtained in stratum II is in the<br />
range of values previously determined for coniferous forest<br />
soils (Myrold 1987) while in stratum I it is significantly<br />
lower, revealing a limited N availability to plants.<br />
From these preliminary results it appears that in stratum<br />
I the low soil temperature recorded affects the soil<br />
biological activity, influencing the biochemical pathways<br />
and reducing the nutrient availability to plants. In stratum<br />
II, less affected by frost conditions, the humification process<br />
carried out by the microbial biomass is more pronounced<br />
and influences the quality of the soil organic<br />
matter. The N availability is higher and may contribute to<br />
explaining the greater tree growth rate.<br />
40
Spatial and temporal observations of seasonal thawing in the Northern<br />
Kolyma lowlands<br />
D. G. Fyodorov-Davydov, V. A. Sorokovikov, A. 1. Kholodov, V. E. Ostroumov, S. V. Gubin and D. A. Gilichinski,<br />
*PI. S. Mergelov, **S. P. Davydov and S. A. Zimov<br />
Soil Laboratory, Institute of Physicochemical and Biological Problems in Soil Science, Russian Academy of Sciences, Pushchino,<br />
Moscow Region, Russia<br />
*Soil Geography Laboratory, lnstifufe of Geography, Russian Academy of Sciences, Moscow, Russia<br />
"'Northeast Science Station, Pacific Ocean institute of Geography, Russian Academy of Sciences, Far East Branch, Chersky,<br />
Russia<br />
Between 1996 and <strong>20</strong>02, 15 sites (100x100 m, Fig. 1)<br />
were established in the Kolyma lowlands within an area of<br />
<strong>25</strong>0x300 km for active-layer monitoring (Brown et al.<br />
<strong>20</strong>00). Eleven of the sites represent different tundra subzones:<br />
arctic (Kuropatochya River), typical (Cape Chukochiy,<br />
Yakutskoe Lake) and southern tundra (Chukochya<br />
River, Alaseya River, Konkovaya River, Segodnia pingo,<br />
Ahmelo Lake). The sites differ also in lithology of parent<br />
materials, which accounts for the differing appearance of<br />
the landscape in various sectors of the region. Observations<br />
started at some sites in 1989.<br />
Figure I. Map of the measurement sites: R12A - Kuropatochya River<br />
(top of the Edoma), R12B - Kuropatochya River (slope of the Edoma),<br />
R13 - Cape Chukochiy (Edorna), R13A - Cape Chukochiy (alas), R14<br />
- Chukochya River, R15A - Konkovaya River (alas), R15B - Konk-<br />
ovaya River (Edoma), R16 Segodnya pingo, R17<br />
- Akhmelo Chan-<br />
nel, R18 - Mountain Rodinka, R19 - Glukhoe Lake, R<strong>20</strong> - Malchikovskaya<br />
Channel, R21 - Ahmelo Lake, R22 - Alazeya River, R<strong>25</strong> -<br />
Yakutskoe Lake.<br />
Ice-rich sediments of the Edoma formation (loess-icy<br />
complex) are widespread in these northwestern lowlands.<br />
These deposits were subjected to intensive thermokarst<br />
development during the Holocene optimum, which has resulted<br />
in the creation of deep depressions or alases.<br />
Thermokarst, alongside with hydraulic erosion, creates<br />
high relief with considerable changes in elevation between<br />
watersheds (edomas), alases and river terraces. Moundy<br />
micorelief is common on watersheds and slopes. Vegetation<br />
is represented by low shrub-grass-moss, grass-mossdryas<br />
or cottongrass-moss-osier associations. The soil<br />
cover consists of complexes with typical cryozems [Glacic<br />
Haploturbels, Typic Haploturbels] and peaty-weakly gleyic<br />
soils [Glacic Aquiturbels, Typic Aquiturbels]. In alas Valleys<br />
or depressions, various polygonal bogs with peatyduff<br />
[Typic Aquorthels], peaty-gleyic, peat-gley [Typic<br />
Aquorthels] and peat [Sphagnic Fibristels, Typic Fibristels,<br />
Typic Sapristels] soils are formed.<br />
Ice-poor sand depositions of the Khallerchinskaya tundra<br />
(Ahmelo Lake, Segodnia pingo) were also influenced<br />
by thermokarst processes. Furthermore, they initially occupied<br />
lower hypsometric levels than the Edoma surfaces.<br />
All these processes resulted in considerable peneplaneization<br />
of the surface, a general landscape waterlogging,<br />
with an extremely high abundance of lakes and assoicated<br />
migration. Because of the moundy microrelief, higher<br />
landscape drainage is common in the southern part of the<br />
area. Zonal biogeocenoses on sandy deposits have low<br />
mound-hummocky microrelief, low shrub-lichen associations<br />
and podzolized podbur [Spodic Psammoturbels] as<br />
the zonal soil subtype. Polygonal bogs vegetation is represented<br />
by moss, grass-moss and moss-lichen-grassy<br />
associations, the soil cover consists of peat-duff or peatduff-gleyic<br />
soils [Psammentic Aquorthels].<br />
The northern taiga subzone is represented by Gluhoe<br />
Lake and Mountain Rodinka. Low mound-hummocky<br />
nanorelief, low shrub-lichen larch open forests and weakly<br />
podzolized sandy podbur [Spodic Psammorthels, Spodic<br />
Psammoturbels] are common for the first site; moss-shrub<br />
larch open forests and gleyic loamy cryozem [Typic<br />
Haploturbels] are formed at the second one.<br />
The Ahmelo and Malchikovskaia Channels sites were<br />
established on the floodplains biogeocenoses with high<br />
heterogeneity. These sites cover both drained parts of<br />
river terraces and different floodplain bogs. On river flood<br />
terraces with moundy microrelief, calamagrostis associations<br />
and sod-gleyic soils [Typic Umbriturbels] are developed.<br />
Among floodplain bogs, sedge-cotton grassy plant<br />
41
Fyodorov-Davydov, D. G. et a4.: Spatial and temporal observations of seasonal thawing.. .<br />
associations with peat-gley soils [Typic Aquorthels] and<br />
comarum-sedge associations with peaty-gley soils [Typic<br />
Aquorthels] prevail.<br />
The lithology of the parent sediments leads to significant<br />
differences in the depth of active layer between the<br />
north-east and north-west zonal tundra ecosystems of<br />
Kolyma lowland: Dry sands thaw 2-3 times deeper than<br />
the moist Edoma Ioams. The differences are considerable,<br />
also for the zonal sandy soils (Ahmelo Lake) and soils of<br />
polygonal bogs (Segodnia pingo) (Table 1).<br />
soil thawed to a considerably shallower depth. This tendency<br />
is better expressed for the zonal tundra sites (Cape<br />
Chukochiy, Yakutskoe Lake, Alaseya River, Ahmelo Lake)<br />
and northern taiga (Gluhoe Lake, Mt. Rodinka) biogeocenoses,<br />
and least pronounced for intrazonal ones (Ahmelo<br />
Channel, Segodnia pingo). A relation between active<br />
layer depth and air temperature at the intrazonal sites of<br />
Konkovaya River (alasj and Malchikovskaya Channel i S<br />
not evident.<br />
" - . . . . .-<br />
Table 1. The mean depth of active layers for Kolyma lowland sites.<br />
0-<br />
Sites<br />
Years<br />
199 I 199 I 199 I <strong>20</strong>0 I <strong>20</strong>0 I <strong>20</strong>0<br />
Figure 2. The 1989-<strong>20</strong>02 changes of the maximal active layer thickness<br />
on different microrelief forms (sandy tundra soils at Ahmelo<br />
Lake).<br />
ACKNOWLEDEMENT<br />
This report was developed as a result of the CALM synthesis<br />
workshop held in November <strong>20</strong>02 at the University<br />
of Delaware.<br />
*- top of the edoma<br />
** - alas.<br />
REFERENCES<br />
In tundras of the western part of the Kolyma lowlands,<br />
the climatic zonality is given at weekly intervals but still<br />
leads to some increase in thaw depths from north to south<br />
in soils of watersheds and slopes. In arctic tundras, the<br />
mean long-term active-layer depth varied between 32 and<br />
37 cm, in typical tundras between 36 and 38 cm, and in<br />
southern tundras between 35 and 47 cm. The comparison<br />
of two sites at Cape Chukochiy revealed greater thaw<br />
depths on the upland watersheds than on waterlogged<br />
alas lowlands. At the same time, in the basin of Konkovaya<br />
River, warming effects could be due to a thicker<br />
snow cover in winter on the alas than on the watershed.<br />
The soils of floodplains, both at the boundary between<br />
tundra and taiga (Ahmelo Channel) and in the northern<br />
taiga (Malchikovskaia Channel) differ from zonal soils at<br />
the same latitude with a considerably smaller active layer<br />
depth.<br />
The 1989-<strong>20</strong>02 data do not show any directional trends<br />
in the changes of the active layer thickness (Fig. 2). However,<br />
this relatively long time series indicates the dependence<br />
of active layer depths on summer temperature. The<br />
increase of thaw depth in the last three years (<strong>20</strong>00-02)<br />
corresponds with high average summer (June-August)<br />
temperatures (11.3-12.1oC) at most sites (Table 1). In the<br />
years (1998-99) with lower temperatures (8.3-9.8oC) the<br />
42<br />
Brown, J., Hinkel K. M. and Nelson, F. E. <strong>20</strong>00. The Circumpolar Active<br />
Layer Monitoring (CALM) Program: <strong>Research</strong> Designs and<br />
Initial Results. Polar Geography 24 (3): 165-<strong>25</strong>8.
8th ~ n ~ Conference ~ ~ ~ on ~ Permafrost ~ i ~ Extended ~ a Abstracts l on Currsnl <strong>Research</strong> and Newly Available Information<br />
Landscape indication of permafrost degradation in Central Siberia<br />
S. P. Gorshkov and MA. Tishkova<br />
Moscow State University, Moscow, Russia<br />
One of the systems that is very responsive to climate<br />
warming is the periphery of the permafrost zone, i.e., the<br />
area of high temperature (- 0.10 -- I .OoC) permafrost. It is<br />
important to know what is happening and what will happen<br />
to the permafrost and permafrost landscapes in the coming<br />
decades.<br />
During the last 5-6 years the table of high-temperature<br />
permafrost in the lower reaches of the Stony Tunguska<br />
River has descended I m or deeper. Degradation takes<br />
place on about 90% of the total permafrost area is in that<br />
region.<br />
According to the elaborated classification (Gorshkov, et<br />
al., <strong>20</strong>03) sensitive, or the least resistant, permafrost is<br />
associated with the sites of low top surfaces (altitudes<br />
about <strong>20</strong>0-<strong>25</strong>0 m) and gentle slopes (steepness about 2-<br />
30). The largest areas of such permafrost are situated in<br />
the field of Pleistocene moraine and lacustrine-glacial deposits.<br />
Both are represented by thin sands, aleurite, loam<br />
and clays, sometimes with boulders. These dynamic permafrost<br />
landscapes are swampy, with low sparse (canopy<br />
4040%) cedar pine-spruce taiga forests with some<br />
birches and larches, sometimes with firs, on peaty-gley<br />
permafrost soils with moss-and-underbrush ground cover<br />
and reindeer lichens in places. Solifluction is identified by<br />
the presence of tilted trees (“drunken forest”) and gaps<br />
(so-called “ruptures-windows”) of 1-2 meters in diameter<br />
and 0.5-0.6 meters in depth, which disturb the turf-plant<br />
layer. Usually gaps are filled with water during warm periods<br />
and have a viscous-flow clay mass at the bottom. Almost<br />
all above-mentioned landscape features are also<br />
typical to the areas covered by birch forests. One important<br />
difference is the distribution of trees. Two, three, four,<br />
sometimes five or six and even seven-eight birches from 8<br />
to 15 meters high grow together like big bushes. They alternate<br />
with small wet lawns.<br />
At the end of the summers of 1995-1997, the thickness<br />
of the active layer was not more than 0.8-1.0 m in the<br />
mentioned landscapes. But during the same period of<br />
<strong>20</strong>02 it was about 1.2-1.5 m and sometimes 1.9 m and<br />
deeper. The gaps were mostly without water. The rate of<br />
active layer thickness increase is estimated to be 10-<br />
15 cmlyear during 1998-<strong>20</strong>02. Sites of heat-deficient<br />
slopes, glaces, and glacis floodplains are characterized by<br />
the presence of more resistant permafrost. Under forest<br />
vegetation at these sites we observed the permafrost table<br />
at a depth of 0.5-0.8 m. We suppose that the active layer<br />
thickness increased not more than 0.2-0.3 m during last<br />
four years. More pronounced changes occurred in areas<br />
located near thermokarst ponds and where forests or<br />
dense bush groves are absent. In the first case the per-<br />
mafrost table is located at a depth of I .O and I .3 m under-<br />
neath holes and hummocks. In the second case geocryological<br />
changes have not been so significant.<br />
Three types of permafrost sites are the most resistant.<br />
These include kurums and ((hanging)) peat bogs of heatdeficient<br />
slopes, as well as forested floodplains of major<br />
rivers.<br />
Figure 1. Overgrowing kurum edge on the left slope of the Rybnaya<br />
River valley in its middle reaches<br />
Kurums could probably be transferred to the previous<br />
category (Fig. I). An increasing number of blocks in unstable<br />
positions is typical for kurums. Furthermore, many<br />
blocks and spaces between them are being covered with<br />
lichens. Water, which flowed at the base of the block horizon<br />
has disappeared. The cold air between the blocks became<br />
warmer. lt seems that during the last several years<br />
a lowering of the permafrost table took place at kurum<br />
43
~~<br />
~~<br />
"<br />
Gorshkov, S.P. et al.: Landscape<br />
sites. Perhaps ice disappeared from the base of kurums in<br />
some places.<br />
The geocryological situation remains without apparent<br />
changes at hanging bog sites. The position of the permafrost<br />
table is approximately at the former level: about 0.4-<br />
0.5 meter under holes and 0.7-0.8 m under hummocks.<br />
Permafrost is most stable at sites of boggy forested<br />
flood-plains in the largest river valleys of the region. This<br />
was proved by the investigations carried out in the valleys<br />
of Kulinna, Stolbovaya, Vorogovka and Stony Tunguska<br />
rivers (Fig 2). At these sites the depth of permafrost table<br />
is about 0.2-0.3 m under holes and reaches 0.4-0.5 m<br />
under hummocks.<br />
indication of permafrost ~ ~ ~ .. ~ a ~ a ~ i ~ ~ "<br />
8) signs of noticeable warming thoroughly of underground<br />
kurum space after water supplying decrease;<br />
9) possible lowering of the permafrost table in kurum<br />
stows;<br />
IO) intensification of thermokarst processes;<br />
11) wide development of long-frozen rocks which can<br />
indicate recent permafrost degradation in places;<br />
12) occurrence of young fir-tree stands within burnt areas<br />
previously covered with different coniferous<br />
species.<br />
During the last few years, a considerable increase in<br />
active layer depth has occurred in the permafrost landscapes<br />
of Central Siberia. Degradation takes place on<br />
about 90% of the permafrost area in the region.<br />
Intensification of solifluction and thermokarst is associated<br />
with this process as well as transformation of kurums.<br />
Degradation of high temperature permafrost is possible in<br />
places. However, several types of sites which are most<br />
resistant to global warming do not show clear responses<br />
to current climate change. Nevertheless the abovementioned<br />
processes, particularly solifluction, can be considered<br />
as "trigger" mechanisms representing a largescale<br />
response to global warming in the studied region.<br />
ACKNOWLEDGEMENT<br />
Figure 2. Floodplain under birch taiga on peaty-boggy permafrost soils<br />
on the left bank of the Stolbovaya River in its lower reaches. Several<br />
trees have been tilted under the impact of ice drift during the flood in<br />
the spring of <strong>20</strong>02.<br />
Thus the following processes can be singled out as<br />
likely rewonses to climate warmina:<br />
I- ~<br />
increase of the active layer thickness and intensification<br />
of solifluction processes;<br />
local replacement of solifluction processes by<br />
landslide mass movement in the areas of active<br />
river erosion;<br />
frequent falling of trees with creeping root system.<br />
This usually happens where the hydromorphic<br />
clays having a viscous-plastic consistence reach a<br />
thickness of about I .5 m or more<br />
presence of trees with bent trunks and vertical tops;<br />
improvement of drainage on top surfaces and<br />
adjacent gentle slopes (water disappeared from<br />
rupture-windows; it is often absent in fissuresruptures);<br />
increase in the number of blocks in unstable<br />
positions on the surface of kurums as well as of<br />
size and number of reindeer moss patches;<br />
exhaustion of ground rills under block cover of ku-<br />
This research is carried out within the framework of RFBR<br />
project 05-01 -64901.<br />
REFERENCE<br />
Gorshkov, S., Vandenberghe, J., Alexeev, B., Mochalova, 0. and<br />
Tishkova, M. <strong>20</strong>03. Climate, permafrost and landscapes of middle<br />
yenisei region. Faculty of Geography of M.V.Lomonosov Moscow<br />
State University, Moscow, (In Russian and in English).
8th International Conference on Permafrost Extended Abstracts an Current <strong>Research</strong> and Newly Available ~ n ~<br />
New permafrost studies at Lake Hovsgol, north-central Mongolia<br />
C. Goulden, *N. Sharkhuu, Ya. Jambaljav, **B. Etzelmiiller, E. S.F. Heggem, ***V. Romanovsky,<br />
****R. Frauenfelder, A. Kaab, *****l. Ariuntsefseg, N. Saruul and S. Tumentsefseg<br />
Mongolian Institute, Academy of Natural Sciences, Philadelphia, U.S.A.<br />
*Institute of Geography, Mongolian Academy of Sciences, Ulaan Baatar, Mongolia<br />
**Department of Physical Geography, University of Oslo, Oslo, Norway<br />
***Geophysical Institute, University of Alaska, Fairbanks, U. S.A.<br />
****Glaciology and Geomorphodynamics Group, Department Geography, University of <strong>Zurich</strong>, <strong>Zurich</strong>, <strong>Switzerland</strong><br />
*****HovsgoI GEF project, Geoecology institute, Mongolian Academy of Sciences, Ulaan Baatar, Mongolia<br />
Lake Hovsgol in northern Mongolia occupies a north-south<br />
aligned tectonic basin at the southern end of the Baikal<br />
Rift System at an elevation of 1645 m, in a biogeographic<br />
and ecological transition zone from taiga forest to steppe<br />
and at the southern edge of the continuous permafrost<br />
zone (Fig. 1). Hovsgol is a Mongolian Long Term Ecological<br />
<strong>Research</strong> network site (Goulden et al. <strong>20</strong>00). In<br />
<strong>20</strong>02 a research program funded by a five-year grant from<br />
the Global Environment Facility to the Geoecology Institute<br />
of the Mongolian Academy of Sciences was initiated<br />
to define the impacts of deforestation, nomadic pasture<br />
use, and climate change on soil and permafrost conditions,<br />
and plant and animal diversity. Analysis of long-term<br />
temperature data from the Hatgal meteorological station at<br />
the south end of the Lake indicates that annual temperatures<br />
have warmed by about 1.5 "C since 1963.<br />
Figure 1. Key map and study area of Lake Hovsgol GEF monitoring<br />
and research valleys. The area lies on the eastern shore of the lake.<br />
Six valleys along the northeastern shore of the Lake<br />
have been selected for study, ranging from the heavily<br />
grazed northern valleys, Turag (51" 17'46" N,<br />
lOO"47'51"E) and Shagnuul, south of Hanh, Noyon and<br />
Sevsuul with moderate grazing, and Dalbay and Borsog<br />
(50°58'24"N, IOO"44'54"E) valleys in the south with little<br />
N<br />
or no grazing pressure (Fig. I). This gradient of pastoral<br />
use allows us to define impacts of climate change in the<br />
south, and its combined effects with pastoral use in the<br />
north on biodiversity and ecosystems. The valleys have a<br />
similar alignment, flowing from east to west, with northfacing<br />
forested slopes covered by larch (Larix sibirica) and<br />
south-facing mixed forest and steppe slopes and steppe<br />
valley bottoms. Two meteorological stations monitor climate<br />
conditions, one in Dalbay Valley, and the second at<br />
Hatgal.<br />
The goals of the permafrost studies are to define the distribution<br />
and temperature of permafrost, the depth of active<br />
layers in detail, and begin a monitoring program of soil<br />
and permafrost temperatures in each of the valleys. In<br />
June <strong>20</strong>02, N. Sharkhuu drilled six boreholes to a depth of<br />
six meters at different topographic locations in the Borsog<br />
and Dalbay valleys. In late August to early September<br />
<strong>20</strong>02, studies of the distribution of permafrost and active<br />
layer depths were initiated, and temperature data loggers<br />
were placed in two watersheds to add to information<br />
gained from boreholes for monitoring soil and permafrost<br />
temperatures in the study area (Fig. 1, Etzelmuller et al.<br />
<strong>20</strong>01).<br />
Sandy soils dominate on slopes but thick sequences of<br />
lake sediment layers cover the valley bottoms. 1 D resistivity<br />
soundings were carried out at four sites, and 2D resistivity<br />
tomography at ten sites in the field. Resistivities were<br />
generally low, calibration with borehole temperatures indicate<br />
permafrost conditions at resistivities close to I kQm,<br />
while ice-rich sediments in the valley bottoms showed resistivities<br />
up to 10 kQm. The results have shown that<br />
permafrost occurs on north-facing slopes and in valley<br />
bottoms. Permafrost thickness in the valley bottoms is estimated<br />
to be between IOand 15 m, while active layer<br />
thickness was around 3-5 m on dry locations and down to<br />
below 1 m in wet, vegetation-covered sites. South-facing<br />
slopes have little or no permafrost. In valleys where intense<br />
grazing occurs, soils are very dry, resulting in relatively<br />
high resistivities even in the active layer. In these<br />
45
Goulden, C. et al.: Permafrost studies, north-central Mongolia ...<br />
/<br />
a a d t h i c k a c a a<br />
Figure 2. Preliminary permafrost profile of Borsog and Dalbay valleys, Lake Hovsgol, northern Mongolia (from N. Sharkhuu). For an overview of<br />
recent permafrost changes in Mongolia, see Sharkhuu (<strong>20</strong>03).<br />
areas, more borehole information is required; additional<br />
boreholes will be drilled in early <strong>20</strong>03 (Fig. 2).Two MRC<br />
soil temperature measurement probes were installed at<br />
the two meteorological stations, one on the north-facing<br />
slope in the Dalbay River valley, and the other at the Hatgal<br />
meteorological station. These temperature probes will<br />
provide year-round hourly soil temperature data at eleven<br />
different depths (approximately every IOcm), starting at<br />
the ground surface and down to one meter. These data,<br />
together with temperature data from the deeper boreholes,<br />
will be used for calibration of a numerical thermal model<br />
developed by the Permafrost Lab, Geophysical Institute,<br />
UAF (Romanovsky et al. 1997, Romanovsky and Osterkamp<br />
<strong>20</strong>01). The site-specific calibrated models will be<br />
applied to the entire period of available meteorological<br />
data from this region to reconstruct the permafrost temperature<br />
dynamics within different landscape units.<br />
Transects for the study and monitoring of plant taxa<br />
and biomass growth, terrestrial insects, birds, mammals,<br />
and stream flow, water chemistry and aquatic biology<br />
have been set across each valley and longitudinally along<br />
the streams. Detailed studies in Borsog Valley show that<br />
increased soil moisture occurs on lower slopes associated<br />
with permafrost and plant and insect taxa and biomass<br />
vary with soil moisture. In Turag and Shagnuul with very<br />
dry soil, grazing may have affected permafrost conditions,<br />
plant biomass is low but plant taxa number is high apparently<br />
due to an increase in the number of xerophytic plant<br />
taxa. Experimental studies will begin in <strong>20</strong>03 to study the<br />
impact of plant cover on soil and permafrost temperature.<br />
REFERENCES<br />
Etzelmuller, B., Holzle, M., Soldjorg Flo Heggem, E., Isaksen, K.,<br />
Mittaz, C., Vonder Muhll, D., mdegard, R. S., Haeberli, W. and<br />
Sollid, J. L. <strong>20</strong>01. Mapping and modeling the occurrence and distribution<br />
of mountain permafrost. Norsk Geografisk Tidskrift, 55(4):<br />
186-1 94.<br />
Goulden, C. E., J. Tsogtbaatar, Chuluunkhuyag, W. C. Hession, D.<br />
Tumurbaatar, Ch. Dugarjav, C. Cianfrani, P. Brusilovskiy, G.<br />
Namkhaijantsen and R. Baatar. <strong>20</strong>00. The Mongolian LTER:<br />
Hovsgol National Park. Korean J. Ecol. 23:135-140.<br />
Romanovsky, V.E., Osterkamp, T.E. and Duxbury, N. 1997. An<br />
evaluation of three numerical models used in simulations of the<br />
active layer and permafrost temperature regimes, Cold Regions<br />
Science and Technology. 26 (3): 195-<strong>20</strong>3.<br />
Romanovsky, V. E. and Osterkamp, T. E. <strong>20</strong>01. Permafrost: Changes<br />
and Impacts, in: R. Paepe and V. Melnikov (eds.), Permafrost Response<br />
on Economic Development, Environmental Security and<br />
Natural Resources, Kluwer Academic Publishers: 297-315.<br />
Sharkhuu, N. in press. Recent changes in the permafrost of Mongolia.<br />
<strong>Proceedings</strong>, 8th International Conference on Permafrost. <strong>Zurich</strong>,<br />
<strong>Switzerland</strong>.<br />
46
8th ~ nt~r~~~i~nal Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available l ~ f ~ ~ ~ a ~ l ~ n<br />
Snow-free ground albedo derived from DAIS 7915 hyperspectral imagery<br />
for energy balance modeling in high-alpine topography<br />
S. Gruber, *D. Schlapfer and M. Hoelzle<br />
Glaciology and Geomorphodynamics Group, Department of Geography] University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
*Remote Sensing laboratories, Department of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
INTRODUCTION<br />
In alpine topography, air temperature, solar radiation,<br />
snow cover height and duration exhibit a strong lateral<br />
variability. Statistical models (e.g. Gruber and Hoelzle<br />
<strong>20</strong>01) are only partially able to approximate these processes<br />
with a limited number of input variables such as<br />
mean annual air temperature and simulated solar irradiation.<br />
Physically-based distributed modeling of the vertical<br />
energy balance can realistically simulate complex interactions<br />
of several variables as demonstrated by Stocker-<br />
Mittaz et al. (<strong>20</strong>02). Their model PERMEBAL simulates<br />
the daily vertical energy balance, based on ground characteristics<br />
and meteorological time series. The required<br />
base data such as albedo, emissivity and roughness<br />
length is needed for every grid cell and must therefore be<br />
measured or accurately estimated. During snow-free conditions<br />
in summer when energy input to the ground is<br />
largest, the daily albedo mean is nearly constant. Remote<br />
sensing is the only way to provide corresponding spatial<br />
data fields. This study uses imaging spectrometry together<br />
with geo-atmospheric correction techniques to measure<br />
albedo in a complex topography. The test site is the<br />
Corvatsch area, Upper Engadin, <strong>Switzerland</strong>, where the<br />
model PERMEBAL was developed.<br />
METHODS AND MEASUREMENTS<br />
Albedo is defined as the ratio of outgoing over incoming<br />
radiation and, for climatological purposes, it is usually<br />
considered in the wavelength interval of about 300 to <strong>25</strong>00<br />
nm, outside of which solar irradiation is small. For periglacia1<br />
surfaces without snow cover, albedo usually lies between<br />
0.1 and 0.3. In the determination of albedo with remote<br />
sensing, ground reflected radiance as well as<br />
irradiance needs to be quantified. As a consequence, a<br />
well-known and well-calibrated sensor is needed together<br />
with accurate characterization of the atmosphere and solar<br />
illumination geometry. Because of the high spatial variability<br />
of surface characteristics and illumination, all auxiliary<br />
and image data must be accurately referenced.<br />
In a raw image, sun-exposed slopes will generally appear<br />
brighter than other areas. Haze in a valley may<br />
lighten-up the valley bottom. However, these differences<br />
largely reflect the influence of atmospheric and illumination<br />
conditions but not the surface property under investigation.<br />
The geo-atmospheric correction needs to account<br />
for this effect and convert an at-sensor radiance image to<br />
reflectance. However, this reflectance (Le. hemisphericaldirectional<br />
reflectance factor, HDRF) is still not suitable to<br />
describe albedo because the reflectance of natural objects<br />
usually exhibits a pronounced angular anisotropy. It is<br />
mathematically described by the bi-directional reflectance<br />
distribution function (BRDF). Multiangular observations of<br />
a target and the use of a BRDF model is necessary to<br />
quantitatively take into account angular anisotropy. Alternatively,<br />
BRDF effects can be corrected using empirical<br />
models. Both approaches are used in this study to derive<br />
spectral albedos for each band. Broadband albedo can<br />
then be calculated by spectral integration.<br />
On 14 August <strong>20</strong>02, the test area was covered by<br />
three flight lines of the imaging spectrometer DAIS7915<br />
under cloud-free conditions. The airborne DAIS7915 sensor<br />
has 72 bands between 500 and <strong>25</strong>00 nm and seven<br />
additional infrared bands.<br />
PROCESSING AND RESULTS<br />
The system calibrated data has been geometrically corrected<br />
using the software PARGE (Schlapfer & Richter,<br />
<strong>20</strong>02). Aircraft position and attitude as well as a digital<br />
elevation model are used to reconstruct the ground-sensor<br />
geometry and precisely locate individual pixel footprints.<br />
The final pixel size is 10 m and the standard errors in horizontal<br />
accuracy were below 5 m. Atmospheric correction<br />
has been performed for the image bands between 500<br />
and <strong>25</strong>00 nm wavelength using the software package AT-<br />
COR4 (Richter and Schlapfer <strong>20</strong>02). The effects of path<br />
radiance, transmittance and direct and diffuse solar flux in<br />
complex topography are accounted for.<br />
In one experiment, the BRDF effect is empirically corrected<br />
based on illumination geometry and using the algo-<br />
47
ithm provided by ATCOR4. The correlation between solar<br />
illumination intensity and the derived albedo serves as a<br />
means to evaluate the removal of topographic effects as,<br />
in theory, they should be virtually independent. Figure 1<br />
demonstrates the success in removing the effects of illumination<br />
and BRDF. The analysis is based upon 56,000<br />
pixels of periglacial surfaces without snow. During this<br />
empirical correction, however, the quantifiable physical<br />
basis for the albedo product is lost. A comparison with<br />
field measurements from one site gives good agreement,<br />
280 pixels on the Murtel rock glacier had an albedo of<br />
17.1 k 1.4% and the snow-free albedo derived from a<br />
CM3 pyranometer (0.3 - 3 um) is 17% (Stocker-Mittaz et<br />
al. <strong>20</strong>02).<br />
W<br />
0<br />
a,<br />
P<br />
0.6<br />
0.5<br />
0.4<br />
E 0.3<br />
m<br />
E<br />
0.2<br />
0.1<br />
0.0<br />
0.0 0.2 0.4 0.6 0.8 1 .o<br />
cosine of solar illumination angle<br />
Figure 1. Dependence of albedo on solar illumination without topographic<br />
correction (thin line), with topographic correction (crosses) and<br />
with topographic correction and empirical BRDF compensation (thick<br />
line).<br />
In a second experiment (see Gruber et al. <strong>20</strong>03 for<br />
details) a BRDF model was fitted to the data and the effects<br />
of angular anisotropy were explicitly corrected. The<br />
parameterization of albedo by a BRDF model provided the<br />
opportunity to investigate the influence that angular anisotropy<br />
has on the net short-wave radiation in a complex terrain.<br />
This was achieved by coupling the energy-balance<br />
model PERMEBAL to a BRDF-based model of albedo.<br />
Using meteorological data of the Corvatsch area, direct<br />
and diffuse downwelling short-wave, radiation as well as<br />
solar incidence angles were parameterized hourly for the<br />
entire year 1999. Slope angles between 0 and 50" northand<br />
south facing were simulated. The solar incidence angle<br />
and the ratio of diffuse and direct radiation served as<br />
input for the parameterization of albedo. The diurnal variation<br />
of albedo measured by the climate station could successfully<br />
be reproduced using this combination of remote<br />
sensing data and PERMEBAL. Unfortunately, the accuracy<br />
of the derived albedo is between 15 and <strong>25</strong>%, only. It<br />
was demonstrated that the same surface material has a<br />
distinctly different albedo on slopes of different orientation<br />
due to the dependence of albedo on solar incidence angles.<br />
This difference is up to <strong>25</strong>% and, therefore, an an-<br />
gular parameterization of albedo should be considered in<br />
complex terrain for accurate models of net short-wave radiation.<br />
CONCLUSION AND OUTLOOK<br />
The derivation of accurate albedo in complex terrain is still<br />
subject to large errors, especially because of the difficulty<br />
to quantify the BRDF for each pixel. The dependence of<br />
albedo on solar incidence angle can lead to distinctly different<br />
albedo values for the same surface in differing topographic<br />
situations. This makes research into angular<br />
anisotropy of albedo especially interesting for energybalance<br />
modeling in complex topography. The variability<br />
ranges of albedo as well as the achieved benefit to model<br />
accuracy can now be assessed based on the data generated<br />
in this project. The two BRDF corrections using an<br />
empirical model and a BRDF model can be compared.<br />
Furthermore, the data generated in this campaign will be<br />
used to assess the error inherent in derivations of albedo<br />
from operational orbiting satellites such as Landsat, SPOT<br />
or ASTER.<br />
ACKNOWLEDGEMENTS<br />
This work has been supported by the Swiss National Science<br />
Foundation project "Analysis and Spatial Modelling<br />
of Permafrost Distribution in Cold-Mountain Areas by Integration<br />
of Advanced Remote Sensing Technology". The<br />
EU-funded project HySens gave additional support.<br />
REFERENCES<br />
Gruber, S. and Hoelzle, M. <strong>20</strong>01. Statistical Modelling of mountain<br />
permafrost distribution I local calibration and incorporation of remotely<br />
sensed data. Permafrost and Periglacial Processes 12(1):<br />
69-77.<br />
Gruber, S., Schlaepfer, D. and Hoelzle, M. <strong>20</strong>03. Imaging Spectrometry<br />
in High-Alpine Topography: The Derivation of Accurate<br />
Broadband Albedo. Proc. the 3rd EARSeL Workshop on Imaging<br />
Spectroscopy, Oberpfaffenhofen, May 13-16 <strong>20</strong>03.<br />
Richter, R. and Schlapfer, D. <strong>20</strong>02. Geo-atmospheric Processing of<br />
Airborne Imaging Spectrometry Data Part 2: atmosphericltopographic<br />
correction. International Journal of Remote<br />
Sensing 23: 2631-2649.<br />
Schlapfer, D. and Richter, R. <strong>20</strong>02. Geo-atmospheric Processing of<br />
Airborne Imaging Spectrometry Data Part 1 : Parametric Orthorectification.<br />
International Journal of Remote Sensing 23: 2609-2630.<br />
Stocker-Mittaz, C., Hoelzle, M. and Haeberli, W. <strong>20</strong>02. Modelling alpine<br />
permafrost distribution based on energy-balance data: a first<br />
step. Permafrost and Periglacial Processes 13(4): 271-282.<br />
48
Permafrost conditions in the starting zone of the Kolka-Karmadon rocklice<br />
slide of <strong>20</strong> September <strong>20</strong>02 in North Osetia (Russian Caucasus)<br />
W. Haeberh, C. Huggel, A. Kaab, *A. Polkvoj, "*/. Zotikov and **N. Osokin<br />
Glaciology and Geomorphodynamics Group, Deparfment of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
*Geological Survey, Environment Department, North Osetia, Vladikavkas<br />
"Russian Academy of Sciences, Moscow<br />
On the evening of <strong>20</strong> September <strong>20</strong>02, a large rocklice<br />
slide took place in the norh-northeast wall of the summit of<br />
Dzhimarai-khokh, northern slope of the Kazbek massif,<br />
Northern Osetia, Russian Caucasus. Large parts of the<br />
debris covered Kolka Glacier at the foot of the slope<br />
sheared off as a consequence of the main impact and<br />
enlarged the initially detached mass by about 60 million<br />
m3 of snowlfirnlice, debris, morainic material and water.<br />
After sliding down at an average speed of about <strong>20</strong>0 kmlh<br />
along a path of 18 km, the moving mass was strongly<br />
compressed at the narrow entrance of the Genaldon<br />
gorge (ca. 1300 m as.l.), ejecting a mudflow which continued<br />
for another 15 km into the Giseldon valley. The<br />
compressed avalanche deposit dammed local rivers and<br />
led to the formation of several lakes. The largest of these<br />
lakes inundated parts of the village of Saniba at the orographic<br />
right valley side near Karmadon and constituted a<br />
growing threat to the population of the downvalley town of<br />
Gisel (Kaab and others <strong>20</strong>03, Popovnin and others in<br />
press).<br />
The situation necessitated the immediate establishment<br />
of an observationlalarm system at the dammed<br />
lakes in order to protect downvalley settlements from potential<br />
floods and avoid unnecessary evacuation or uncontrolled<br />
reactions of inhabitants. At the same time, the<br />
potential for an immediate repetition of similar or even<br />
larger accidents from the mountain slope had to be assessed.<br />
This involved interpretation of photographs collected<br />
by the authorities during reconnaissance flights by<br />
helicopter together with some best-guesses about thermal<br />
aspects related to firn, ice and permafrost conditions in the<br />
starting zone. The fact that two large slides from the same<br />
slope but with considerably shorter runout distances had<br />
occurred in 1902 - almost exactly 100 years ago - constituted<br />
an important aspect of this assessment.<br />
The avalanche starting zone as depicted in Figure I<br />
was approximately 1 km wide and was located between<br />
about 4300 m and 3500 m a.s.1. A roughly estimated total<br />
of some 4 million m3 of steeply inclined metamorphic rock<br />
layers were detached to a depth of about 40 m, entraining<br />
a similar thicknesslvolume of snow, firn and glacier ice.<br />
The dip of the bedrock layers, as visible in the detachment<br />
zone after the event, was less than the inclination of the<br />
now-exposed slope surface: outcropping of steeply inclined<br />
bedrock layers in the lower part of the affected<br />
slope (arrow in Fig. 1) could have constituted an important<br />
geological influence on stability conditions. The primary<br />
cause of the instability must have been within bedrock<br />
rather than surface ice. As bedrock stability can be especially<br />
low in warm or degrading permafrost (Davies and<br />
others <strong>20</strong>01), thermal conditions affecting ice and water<br />
within rock fissures are likely to have exerted major and<br />
detrimental influence.<br />
Figure 1. Starting zone of the Kolka-Karmadon rocWice slide in the<br />
north-northeast wall of the summit of Dzhimarai-khokh. Arrow points<br />
to steeply inclined bedrock layers.<br />
A weather station near Karmadon was run between<br />
1962 - 1987, showing a mean annual air temperature<br />
(MAAT) of 59°C. Assuming a warmer recent MAAT of<br />
about 7°C at 1300 m a.s.l., and a MAAT-lapse rate of 0.6<br />
f O.I"CllOOm, present-day mean annual air temperature<br />
can be estimated at -6 f 2°C at the lower and -11 f 3°C<br />
at the upper end of the detachment zone. The existence of<br />
active rock glaciers in the region (Fig. 2) indeed confirms<br />
that the lower boundary of discontinuous mountain per-<br />
49
Haeberli, W. et al.: Permafrost conditions in the starting zone of the Kolka-Karmadon rocklice slide ...<br />
mafrost is around <strong>25</strong>00 m a.s.l., where MAAT is slightly<br />
below 0°C. Direct minilogger measurements in the Alps<br />
(Gruber and others <strong>20</strong>03) indicate that mean annual rock<br />
temperatures tend to be somewhat higher than air temperatures<br />
extrapolated from weather stations. This effect,<br />
however is to some degree compensated for by the fact<br />
that the affected slope is oriented away from the sun and<br />
hence will be colder than average. In the absence of any<br />
direct measurements or numerical model results, bedrock<br />
surface temperatures in the detachment zone may be estimated<br />
at about -5 to -1 O T , indicating bedrock conditions<br />
of cold permafrost after the event.<br />
tioned by reconnaissance teams, documents that thermal<br />
water surfaces in the critical zone.<br />
The answer to the question about the potential for an<br />
immediate repetition of a similar or even larger event was<br />
based on the facts that (I) the elimination of hanging glaciers<br />
with warm firn areas had reduced the load on the<br />
slope and eliminated the main meltwater source, (2) the<br />
exposed bedrock would now be subject to strong cooling<br />
and deep freezing, (3) the sheared-off iceldebris volume<br />
of Kolka glacier had made large mass increase at the foot<br />
of the slope impossible and (4) the path of the <strong>20</strong> September<br />
event with its heavy destruction was practically inaccessible<br />
anyway. It was concluded that the threat from a<br />
possible repetition at the same site and in the immediate<br />
future of an even bigger accident could be considered<br />
minimal but that instabilities in the remaining parts of the<br />
slope could continue and should be observed accordingly.<br />
The build-up over long time intervals of hanging glaciers<br />
with complex thermal conditions, in combination with<br />
thermal water in the region and the specific geological<br />
setting (steeply inclined rock layers, volcaniclgeothermal<br />
processes), could explain the "quasiperiodicity" of events<br />
at a century time scale. Effects of potential future atmospheric<br />
warming would also have to be taken into account<br />
for the coming decades.<br />
ACKNOWLEDGEMENT<br />
Figure 2. Upper avalanche path of the Kolka-Karmadon rock/ice slide,<br />
the summit of Dzhimarai-khokh and the detachment zone are in the<br />
background. An active talus-derived rock glacier (arrow) is seen in the<br />
lower right corner. Note steamldust () cloud at the foot of the slope<br />
where Kolka glacier has been sheared off (star) and overriden tongue<br />
of Maili glacier (circle). Kazbek volcano would be to the left.<br />
During the years before the event, the slope had been<br />
covered by a number of well-developed, spectacular<br />
hanging glaciers. Such firnlice bodies are known to have a<br />
complex distribution of englacial temperatures: while verticallimpermeable<br />
cliffs, where ablation takes place through<br />
ice breaking off, can be as cold as the surrounding bedrock,<br />
permeable firn layers underneath the less inclined<br />
upper surface with snow accumulation are strongly<br />
warmed up by latent heat from percolating and refreezing<br />
meltwater. In areas with MAAT exceeding -10 to -12"C,<br />
firn is usually temperate and the icelbedrock interface behind<br />
the cold cliff remains at phase equilibrium temperature,<br />
a pattern, which introduces deep-seated thermal<br />
anomalies within the underlying bedrock (Haeberli and<br />
others 1997). It is, therefore, highly probable that the detachment<br />
zone at Dzhimarai-khokh was in a complex condition<br />
of relatively coldlthick permafrost combined with<br />
warm if not unfrozen parts and meltwater flow in very<br />
steeply inclined materials with most heterogenous permeability.<br />
This situation favouring high and strongly variable<br />
water pressures was further complicated by the fact that<br />
hot springs are known to occur in this volcanic region of<br />
the Kazbek massif: the steamldust () clouds visible on<br />
most photographs, and the strong sulphur odour men-<br />
The presented work resulted from collaboration with the<br />
the Government of North Osetia and the Academy of Sciences,<br />
Moscow, as part of a project directed and funded<br />
by the Swiss Development Corporation - Humanitarian<br />
Aid.<br />
REFERENCES<br />
Davies, M., Hamza, 0. and Harris, C. <strong>20</strong>01. The effect of rise in mean<br />
annual air temperature on the stability of rock slopes containing<br />
ice-filled discontinuities. Permafrost and Periglacial Processes 12:<br />
137-144.<br />
Gruber, S., Peter, M., Hoelzle, M., Woodhatch, I. and Haeberli, W.<br />
<strong>20</strong>03. Surface temperatures in steep Alpine rock faces - a strategy<br />
for regional-scale measurements and modelling. <strong>Proceedings</strong> of<br />
the 8th International Conference on Permafrost, <strong>Zurich</strong>, <strong>Switzerland</strong>.<br />
Haeberli, W., Wegmann, M. and Vonder Muhll, D. 1997. Slope stability<br />
problems related to glacier shrinkage and permafrost degradation<br />
in the Alps. Eclogae geologicae Helvetiae 90: 407-414.<br />
Kaab, A., Wessels, R., Haeberli, W., Huggel, C., Kargel. J.S. and<br />
Khalsa, S.J.S. <strong>20</strong>03. Rapid Aster imaging facilitates timely assessments<br />
of glacier hazards and disasters. EOS 13/84: 117, 121.<br />
Popovnin, V.V., Petrakov, D.A., Toutoubalina, O.V. and Tchernomorets,<br />
S.S., <strong>20</strong>03. Catastrophe glaciaire <strong>20</strong>02 en Ossetie du<br />
Nord. Unpublished french translation, original article to appear in<br />
Earth Cryosphere VI1 (1).<br />
50
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
The thermal regime of the coarse blocky active layer at the Murtel rock<br />
glacier in the Swiss Alps<br />
S. Hanson and M. Hoelzle<br />
Glaciology and Geomorphodynamics Group, Department of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
Energy balance measurements on the Murtd rock glacier<br />
in the Upper Engadine in the Swiss Alps have shown that<br />
the sum of all measured components indicate a non-zero<br />
energy budget (Mittaz et al. <strong>20</strong>00). This pattern was seen<br />
Haeberli (<strong>20</strong>00) stated that due to pronounced lateral<br />
energy fluxes in inclined active layers consisting of coarse<br />
blocks, pronounced differences can exist between mean<br />
annual temperatures at the surface and at the permafrost<br />
both in 1997 and in 1998 with positive deviations in the table, respectively. Harris and Pedersen (1998) suggest<br />
winter months and negative deviations during the summer.<br />
Already at this time it was suggested that the imbalance of<br />
that the air moving through the voids between the large<br />
blocks play a significant role in modifying the thermal rethe<br />
energy exchange fluxes may be explained by un-<br />
gime of the blocky surface material. They showed, like<br />
measured advective energy fluxes within the active layer.<br />
Furthermore, a de-coupling between the surface tem-<br />
Humlum (1997), how the active layer protects the permafrost<br />
core acting like a thermal filter when snow cover is<br />
perature and the temperature within the active layer of the thin or absent and low wind speeds prevail. The summer<br />
rock glacier was recognized which does support this inter- warming of the permafrost surface is prevented by the<br />
pretation.<br />
trapped cold air between the boulders in the active layer to<br />
the benefit of the permafrost. Furthermore evaporative<br />
cooling in summertime keeps the temperature in the active<br />
layer low, A reverse of these processes caused by an<br />
abrupt increase in air temperatures within the active layer<br />
has also been reported. Humlum (1997) pointed out the<br />
scenario with high wind speed and absence of snow cover<br />
where forced ventilation caused by wind pumping disturbs<br />
the pattern and an introduction of relatively warm air can<br />
Figure 1. The Murtel rock glacier. The square indicates the position of<br />
the measuring site. Photo: C. Rothenbuehler<br />
The existence of permafrost in the Alps is strongly influenced<br />
by incoming short-wave radiation and snow<br />
cover thickness and durations. Prediction of present and<br />
future permafrost distribution in the Swiss Alps is based<br />
mainly on the prediction of the extent of these factors.<br />
However, the above-mentioned result may indicate processes<br />
related to permafrost conditions, especially in<br />
coarse blocky material, that are not fully accounted for today<br />
and could be important in coupling the atmosphere to<br />
the ground when applying climate change scenarios.<br />
have a deepening effect on the active layer. Latent heat<br />
release caused by refreezing of meltwater during spring<br />
can possibly also play an important role in heating up the<br />
active layer.<br />
The importance of each of these non-conductive processes<br />
is highly dependent upon local site conditions and is<br />
also highly variable. Many of the non-conductive processes<br />
are seasonal and effective only during periods of<br />
phase changes when the driven gradient near the ground<br />
surface is relatively large. In the context of global warming,<br />
a better understanding of the non-conductive proc-<br />
esses in the active layer of alpine permafrost and how<br />
these are related to seasonal alterations is required in order<br />
to comprehend and anticipate the influence of climate<br />
changes in the Alps.<br />
The year <strong>20</strong>02 featured two very different cold periods<br />
with snow cover. The winter of <strong>20</strong>01/<strong>20</strong>02 was very cold<br />
and dry. The snow did not arrive before late January <strong>20</strong>02<br />
and the snow cover was thin and discontinuous. This led<br />
to open voids between the boulders during the whole<br />
winter, allowing free connection between the atmosphere<br />
51
and the active layer. The winter of <strong>20</strong>02/<strong>20</strong>03 started already<br />
in October <strong>20</strong>02 with heavy snowfall that covered<br />
the blocky surface completely with several meters of snow<br />
that lasted the year out. The thick snow cover sealed off<br />
any exchange between the active layer and the atmosphere.<br />
These two situations give an unique opportunity to<br />
compare two extreme conditions on the snow cover’s influence<br />
on the processes governing the energy fluxes in<br />
the active layer.<br />
REFERENCES<br />
Haeberli, W. <strong>20</strong>00. Modern research perspectives relating to permafrost<br />
creep and rock glaciers: a discussion. Permafrost and<br />
Periglacial Processes 11 : 290-293.<br />
Harris, S. A. and Pedersen, D. E. 1998. Thermal regime beneath<br />
coarse blocky material. Permafrost and Periglacial Processes 9:<br />
107-1<strong>20</strong>.<br />
Humlum, O., 1997: Active layer thermal regime at three rock glaciers<br />
in Greenland. Permafrost and periglacial processes, 8: 383-408.<br />
Mittaz, C., Hoelzle, M. and Haeberli, W. <strong>20</strong>01. First results and interpretation<br />
of energy-flux measurements of Alpine permafrost. Annals<br />
of Glaciology 31: 275-280.<br />
Figure 2. The energy balance climate mast on the Murtbl rock glacier.<br />
Photo: M. Hoelzle.<br />
The aim of this project is to discuss the thermal regime<br />
of an active layer at a high resolution in a coarse bouldery<br />
environment in an alpine discontinuous permafrost setting.<br />
A fully-equipped energy balance station has been logging<br />
data every 30 minutes at the Murtel rock glacier during the<br />
entire year of <strong>20</strong>02. In the same period, seven thermistors<br />
placed within the active layer of the rock glacier were<br />
measuring temperatures at a 5-minute scale. Two of these<br />
thermistors were measuring the air temperature between<br />
the blocks. The remaining five thermistors were drilled into<br />
the blocks. With the comparison of the meteorological observations<br />
and the recorded temperature within the active<br />
layer, a discussion is presented and data provided of the<br />
processes that govern the state of the active layer. Indications<br />
of non-conductive processes are analyzed in detail<br />
with the temperature gradient within the active layer, and<br />
the cumulative daily heat accumulation at the surface and<br />
within the active layer itself.<br />
52
8th International Conference on Permafrost Extended Abstracts an Current <strong>Research</strong> and Newly Available Information<br />
Detection of permafrost structure by transient electromagnetic methoc 1 in<br />
Mongolia<br />
K. Harada, *K. Wada and **M. Fukuda<br />
Miyagi Agricultural College, Sendai, Japan<br />
*Mitsui Mineral Development Engineering Co. , 1 td. , Tokyo, Japan<br />
**/nstitute of low Temperature Science, Hokkaido University, Sapporo, Japan<br />
Field observations were carried out in order to examine<br />
the applicability of the transient electromagnetic (TEM)<br />
method to detection of permafrost structure in a discontinuous<br />
permafrost area of Mongolia.<br />
In Mongolia, frozen ground occupies nearly 63% of the<br />
total territory of this country, and the thickness of permafrost<br />
varies between 5 and <strong>20</strong>0 m (Choibalsan 1998). The<br />
TEM survey was made in Nalaikh (47O45' N, 107°15' E),<br />
sandstone. Volcanic ash and Quaternary deposit are<br />
above it for approximately 300 m. The soil profile up to 6.8<br />
m is clay and silt confirmed by the boring survey made in<br />
November 1998.<br />
The TEM survey was conducted near a borehole,<br />
named #524. The measuring station interval was 10 m<br />
and the length of the profile was 100 m, named St. 2900 -<br />
St. 3000. St. 3000 was located I00 m from the borehole<br />
-<strong>20</strong> 88 84 76 74 74 67 63 63 63' -<strong>20</strong><br />
~84 84<br />
"r<br />
% -60 ! I<br />
I<br />
-4 0<br />
-60<br />
,<br />
I<br />
-100 - -100<br />
i 700<br />
-1<strong>20</strong> I I I I I I -1<strong>20</strong><br />
7<br />
permafrost<br />
base<br />
I<br />
i<br />
- 1998/11/16<br />
- 1999/6/3<br />
2900 29<strong>20</strong> 2940 2960 2980 3000 -3 -2 -1 0 1 2 3<br />
Distance (m)<br />
Temperature ("C)<br />
Figure 1, (a) Layered earth section near borehole #524 and (b) temperature profile in the borehole, Nalaikh. Values in figure (a)<br />
show resistivity in ohm-m.<br />
which is located in the suburbs of Ulaanbaatar, central #524. A central induction measurement configuration was<br />
Mongolia, in November 1998. Around this area, perma- used in this site. The measurements were performed by<br />
frost is discontinuously distributed and the mean annual using the transmitter loop of 60 x 60 m. The transient data<br />
air temperature ranges between 4 and 0 'C (Choibalsan, were recorded using PROTEM 47 TEM system by Geon-<br />
1998). Bedrock around this area is consisted of schist and ics Limited with a receiver coil having an effective area of<br />
53
31.4 m At the borehole #524, the ground temperature to<br />
the depth of 50 m was measured on 16 November 1998<br />
and 3 June 1999. Based on the temperature profiles, the<br />
permafrost base was located at the depth of 36.4 m (Fig.<br />
1 b).<br />
The resistivity structures near the borehole #524 are<br />
shown in Figure I (a). Three- or four-layered structures<br />
were obtained between St. 2900 and St. 2970. Rather low<br />
resistivity values (40-100 ohm-m) were found, and the<br />
variation of resistivity value was small. The first and third<br />
layers had the lowest resistivity values, 42-62 ohm-m. The<br />
resistivity value of the second layer was higher than that of<br />
the other layers. At the stations between St. 2980 and St.<br />
3000, a two-layered resistivity structure was estimated.<br />
The resistivity values of the first and second layers were<br />
almost constant, 63 and 55 ohm-m, respectively. The difference<br />
of resistivity between the first and second layers<br />
was very small. The lower boundarv of the first layer<br />
ranged from 34 m to 45 m deep.<br />
1<br />
Harada, K. et a!.: Detection af permafrost structure by transient ...<br />
periments (after Harada and Yoshikawa 1996, Harada and<br />
Fukuda <strong>20</strong>00, Harada et al. <strong>20</strong>00). The experimental results<br />
show that the resistivity difference of the fine-grained<br />
soil is small due to the presence of unfrozen water, , especially<br />
in case of low water content. Furthermore, the<br />
measured ground temperature shows above -1 OC below<br />
the depth of 10 m. As a resistivity of frozen soil is a function<br />
of temperature, the resistivity difference becomes<br />
small in this study site.<br />
From the field observation, it is noted that the best-fit<br />
model obtained by the TEM survey represents the permafrost<br />
structure precisely at this site, even though the difference<br />
in resistivity between frozen and unfrozen layers is<br />
small.<br />
REFERENCES<br />
Choibalsan, N. 1998. Characteristics of permafrost and foundation<br />
design in Mongolia. In <strong>Proceedings</strong> of 7th International Conference<br />
on Permafrost: 157-160.<br />
Harada, K. and Fukuda, M. <strong>20</strong>00. Characteristics of the electrical resistivity<br />
of frozen soils. Seppyo 62: 15-22 (in Japanese with English<br />
abstract).<br />
Harada, K., Wada, K. and Fukuda, M. <strong>20</strong>00. Permafrost mapping by<br />
transient electromagnetic method. Permafrost and Periglacial<br />
Processes 11 : 71 -84.<br />
Harada, K. and Yoshikawa, K. 1996. Permafrost age and thickness<br />
near Adventfjorden, Spitsbergen. Polar Geography <strong>20</strong>: 267-281.<br />
-5 0 5<br />
Temperature (OC)<br />
Figure 2. Electrical resistivity of soil samples. q is volumetric<br />
water content (after Harada and Yoshikawa 1996; Harada and<br />
Fukuda <strong>20</strong>00; Harada et al. <strong>20</strong>00). Spitsbergen is the soil sample<br />
obtained from Spitsbergen and CPC is from Caribou-Poker<br />
Creeks, Alaska.<br />
The surface layer represented permafrost as confirmed<br />
by the boring survey. From the observation at the nearest<br />
station of the borehole #524, St. 3000, the lower boundary<br />
of the first layer was located at the depth of 34.2 m from<br />
the best-fit model and the allowable depth ranged between<br />
23 m and 74 m based on the equivalence analysis.<br />
The lower boundary of the best-fit model, 34.2 m, was in<br />
good agreement with the depth of permafrost base obtained<br />
from the ground temperature profiles, 36.4 m.<br />
Therefore, this boundary indicates the permafrost base.<br />
The observational results show that the difference in resistivity<br />
values between the frozen and unfrozen layers<br />
was very small. This small difference may have been<br />
caused by the low water content. Choibalsan (1998) reported<br />
that Mongolian permafrost is generally characterized<br />
by low to moderate ice contents. Figure 2 represents<br />
the relationship between resistivity values of fine-grained<br />
soils and temperature obtained from the laboratory ex-<br />
54
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Application of a capacitively coupled resistivity system for mountain<br />
permafrost studies and implications for the interpretation of resistivity<br />
values<br />
C. Hauck, and *C. Kneisel<br />
Graduiertenkolleg Natural Disasters & lnsfifufe for Meteorology and Climate <strong>Research</strong>, University of Karlsruhe, Germany<br />
“lnstitufe for Physical Geography, Universify<br />
of Wurzburg, Germany<br />
INTRODUCTION<br />
particular frequency by the transmitting dipole. The resulting<br />
voltage coupled to the receiver‘s dipole is measured. In or flat<br />
Geophysical techniques have become more and more easily accessible terrain data collection is faster than when<br />
popular to determine the physical properties of frozen ground, using conventional DC resistivity systems. Further details<br />
as they are comparatively cheap and easy to apply on most are given in Timofeev et al. (1994).<br />
kinds of permafrost terrain. Within the large variety of gee The DC resistivity tomography technique has been used<br />
physical techniques, electric and electromagnetic methods<br />
increasingly in permafrost studies in recent years. The details<br />
are especially well suited for permafrost studies, as resistivity<br />
of the method and the peculiarities of its application to mounincreases<br />
with increasing ice content (or decreasing unfrozen<br />
water content), enabling the determination of ice content and<br />
tain permafrost are described in many recent publications<br />
ground ice characteristics in principle. However, in practical (e.g. Marescot et al. <strong>20</strong>03, Hauck et al. <strong>20</strong>03) and will not be<br />
applications a number of technical limitations often prohibit a repeated here.<br />
reliable assessment of the ice content. Among these, the in- The test sites of this study are located in the Bever valley<br />
fluence of the electrical contact of the electrodes (in DC resis- (eastern Swiss Alps, see Kneisel et al. <strong>20</strong>00) and on a nontivity<br />
surveys) and the influence of data inversion of the frozen grass field near St. Moritz, which was used as refermeasured<br />
apparent resistivities are probably the most impor- ence area. DC resistivity measurements were conducted<br />
tant. Data inversion includes the problem of equivalence, with a SYSCAL Junior Switch system with Wenner and Diwhich<br />
may be addressed by giving ranges for the resistivity pole-Dipole configurations. The OhmMapper surveys were<br />
values of the various layers (e.g. Kneisel 1998), and the in- conducted in Dipole-Dipole configuration using the same array<br />
fluence of inversion parameters, which may change resistiv- geometries as for the SYSCAL surveys.<br />
ity values over up to one order of magnitude (Hauck et al.<br />
<strong>20</strong>03).<br />
In this contribution we would like to address the discrep-<br />
RESULTS<br />
ancy of resistivity results obtained through different measurement<br />
techniques by comparing data from several resistivity<br />
Whereas very similar results were obtained with both methsurveys<br />
along the same profile in the eastern Swiss Alps.<br />
ods on the flat grass field of the non-frozen reference area (not<br />
Hereby, a capacitively coupled resistivity system (Ohmshown<br />
here), difficulties were encountered on the steep and<br />
Mapper, Timofeev et al. 1994), which has been successfully<br />
heterogeneous terrain in the Bever valley. The surface is<br />
applied on Arctic permafrost, is applied on more heterogenevegetated<br />
by small trees and bushes and covered partly by<br />
ous mountain permafrost terrain. The applicability of this new<br />
medium-sized boulders, which makes towing the OhmMapinstrument<br />
was evaluated in comparison to a standard multiper<br />
system a difficult task, as the antenna dipoles tend to get<br />
channel resistivity system.<br />
stuck between boulders and bushes. In addition, downslope<br />
METHODS AND DATA ACQUISITION<br />
The OhmMapper (Geometrics) is a capacitively coupled re<br />
sistivity meter that measures the electrical properties of the<br />
subsurface without the galvanic coupling of ground electrodes<br />
used in traditional resistivity surveys. A simple coaxial-cable<br />
array with transmitter and receiver sections is pulled along<br />
the ground. An alternating current is induced in the earth at a<br />
towing is only manageable if a second person is holding the<br />
end of the line in order to prevent uncontrolled tumbling OT<br />
misalignment of the dipole antennas (which is also the case<br />
on snow-covered slopes).<br />
Figure 1 shows the survey results for both methods in<br />
comparison to the DC resistivity survey conducted in 1998<br />
(Kneisel et al. <strong>20</strong>00). All results are shown as a horizontal<br />
variation of electrical resistivity along the survey line. In principle,<br />
the <strong>20</strong>02 results of the OhmMapper and the DC resistivity<br />
survey (SYSCAL) with Dipole-Dipole configuration<br />
55
Hauck, C. et al.: Application of a capacitivety cuupled rcsistivily system for rnountaine<br />
shouldbethesame,asmeasurementsweretakenon the<br />
same day, at the same location and with the same survey<br />
geometry. Even though the general resistivity distribution is<br />
similar, the OhmMapper results (Fig. la) have much larger<br />
noise content than the SYSCAL data (Fig. I b). This is mostly<br />
due to difficulties in keeping the correct distance between receiver<br />
and transmitter, as well as the correct orientation of the<br />
dipole antennas due to the steep and uneven terrain. Besides<br />
and possibly more important, the absolute values of apparent<br />
resistivities are systematically lower for the OhmMapper (1<br />
kam) than for the SYSCAL (4 kam). In contrast, it is seen<br />
that the differences between the SYSCAL Dipole-Dipole and<br />
Wenner results (Figs. 1 b and IC) and the Wenner results ob<br />
tained in different years (1998 and <strong>20</strong>02, Fig. IC) are less<br />
variable than between OhmMapper and SYSCAL (Figs. la<br />
and 1 b) obtained at the same time and at the same locations. Figure 2. Comparison of measured apparent resistivity values obtained<br />
with SYSCAL and OhmMapper. Values are given in<br />
loglo(Wm).<br />
Figure 1. Comparison of apparent resistivity variation along the<br />
survey line in the Bever valley for (a) OhmMapper, (b) SYSCAL Dipole-Dipole<br />
array for 2 different spacings and (c) Wenner array<br />
from 2 different years.<br />
The discrepancies between SYSCAL and OhmMapper<br />
are larger the higher the resistivity values. Plotting the apparent<br />
resistivity values obtained with the OhmMapper system<br />
against the corresponding apparent resistivity values of the<br />
SYSCAL system, systematic differences are seen for resistivity<br />
values above 500 m with smaller values for the<br />
OhmMapper system (Fig. 2). This systematic discrepancy is<br />
more pronounced the larger the resistivities. For very high<br />
resistivities the OhmMapper values reach less than for the<br />
values obtained with the SYSCAL system.<br />
The possible reason is quite likely found in the dflerent<br />
electrical coupling methods of both techniques. The high contact<br />
resistances typically found in mountain permafrost terrain<br />
result in larger apparent resistivities than obtained with capacitively<br />
coupled instruments (OhmMapper) or electromagnetic<br />
induction systems, where no coupling is needed at all.<br />
In most mountain permafrost applications only relative resistivity<br />
variations are needed, as the aim is to detect lateral w<br />
vertical resistivity anomalies indicating the presence of frozen<br />
material. However, when comparing data from dfferent IOMtions,<br />
care has to be taken as absolute resistivities may be<br />
overestimated due to resistive surface characteristics at specific<br />
locations.<br />
REFERENCES<br />
Hauck, C., Vonder Muhll, D. and Maurer, H. <strong>20</strong>03. Using DC resistivity<br />
tomography to detect and characterize mountain permafrost.<br />
Geophysical Prospecting 51 (in press).<br />
Kneisel, C. 1998. Occurrence of surface ice and ground<br />
icelpermafrost in recently deglaciated glacier forefields,<br />
St.Moritz area, eastern Swiss Alps. Proc. of the 7th Int. Conf. on<br />
Permafrost, Yellowknife, Canada: 575-581.<br />
Kneisel, C., Hauck, C. and Vonder Muhll, D. <strong>20</strong>00. Permafrost below<br />
the timberline confirmed and characterized by geo-electric resistivity<br />
measurements, Bever Valley, eastern Swiss Alps. Permafrost<br />
and Periglacial Processes 11 : 295-304.<br />
Marescot, L., Loke, M.H., Chapellier, D., Delaloye, R., Lambiel, C.<br />
and Reynard, E. <strong>20</strong>03. Assessing reliability of 2D resistivity imaging<br />
in permafrost and rock glacier studies using the depth of<br />
investigation index method. Near Surface Geophysics l(2): 57-<br />
67.<br />
Timofeev, V.M., Rogozinsky, A.W., Hunter, J.A. and Douma, M. 1994.<br />
A new ground resistivity method for engineering and environmental<br />
geophysics. Proc.of the SAGEEP, Annual Meeting: 701-<br />
715.<br />
56
8th International Conference an Permafrost Extended Abstracts an Current <strong>Research</strong> and Newly Available l ~ f ~ ~ ~<br />
Combining and interpreting geoelectric and seismic tomographies in<br />
permafrost studies using fuzzy logic<br />
C. Hauck and *U. Wagner<br />
Graduiertenkolleg Natural Disasters & lnsfitute for Meteorology and Climate <strong>Research</strong>, University of Karlsruhe, Germany<br />
*Graduiertenkolleg Natural Disasters & Institute for Program Structures and Data Organization, University of Karlsruhe, Germany<br />
INTRODUCTION<br />
For many problems in periglacial and glaciological studies,<br />
but also for other environmental, engineering and archaeological<br />
problems, geophysical investigations are the<br />
main source of information concerning the structure and<br />
characteristics of the subsurface. In particular, the ability<br />
to gather 2-dimensional information about the subsurface<br />
material is considered the main advantage of geophysical<br />
techniques as opposed to the single-point information<br />
through boreholes. However, the indirect nature of geophysical<br />
surveys, where material properties like water<br />
content, pore size or temperature have to be inferred from<br />
the measured physical variables such as electrical resistivity,<br />
can be considered a major drawback for the application<br />
of geophysical methods in applied geosciences.<br />
Nevertheless, the uncertainty in the interpretation of geophysical<br />
data sets is seldom explicitly treated in geophysical<br />
applications.<br />
In order to address this uncertainty, a model approach<br />
using fuzzy logic is presented in this contribution, which is<br />
based on expert knowledge about a particular problem.<br />
The problem is taken from the area of ground-ice detection,<br />
where a large variety of geophysical methods has<br />
been applied in recent years (e.g. Vonder Muhll et al.<br />
<strong>20</strong>01, Hauck and Vonder Muhll <strong>20</strong>03). Generally, more<br />
than one method is used in order to get a multivalued data<br />
set and to facilitate data interpretation (e.g., DC resistivity<br />
and refraction seismics). However, due to the nonuniqueness<br />
of the inversion results and the range of possible<br />
values for most earth materials, the interpretation of<br />
the data set can be very ambiguous.<br />
FUZZY MODEL<br />
Fuzzy logic is an extension of a multivalued logic, where<br />
classes of objects have unsharp boundaries and membership<br />
is a matter of degree. It is a convenient way to map<br />
an input space to an output space, especially if a quantitative<br />
output is required from imprecise input variables. In<br />
this case the input space is comprised by the results from<br />
the geophysical surveys (e.g., electrical resistivities or<br />
seismic P-wave velocities) and the output space is the<br />
“degree of ice content” (but not the ice content itself), that<br />
is the possibility of ground-ice occurrence.<br />
Figure 1, Decision surface for the fuzzy inference system for ground<br />
ice detection. Each value of seismic velocity (in kmls) and the logarithm<br />
of electrical resistivity (in loglo(Wm)) is associated with a degree<br />
of possibility for ground ice (z-axis).<br />
The fuzzy inference system used in this study is of the<br />
so-called Mamdani-type (Mamdani and Assilian 1975) and<br />
is based on nine rules, all linking low, medium and high<br />
resistivity and velocity values to a corresponding output<br />
(low, medium and high ice content). A practical view of the<br />
rule system is shown as decision surface (Fig. I), where<br />
each pair of resistivity and velocity data point is associated<br />
with a possibility of ground-ice occurrence. High resistivity<br />
values and medium seismic velocities around 3500 mls<br />
(velocity for pure ice) are associated with a high possibility<br />
of ice occurrence, as can be seen from Fig. 1. Due to the<br />
exponential increase of resistivity with decreasing and<br />
negative temperature, the logarithm of resistivity is used in<br />
the model.<br />
57
RESULTS<br />
Hauck, C. et ai.: C ~ ~ b and ~ interpreting n i ~ ~ geoelectric and seismic tomographies ...<br />
The decision surface and the corresponding fuzzy model<br />
are tested using a synthetic data set of resistivity and<br />
seismic data pairs (Fig. 2). Both data sets are comprised<br />
of 4 regions with high, medium and low resistivity and velocity<br />
values, respectively, where only the upper right hand<br />
corner represents values consistent with the material<br />
properties of ice. Accordingly, the resulting model output<br />
shows a region of high possibility for ice occurrences in<br />
this part, whereas the possibility is low for all other regions.<br />
This outcome would have been difficult to predict by<br />
“manual” inspection of the resistivity and seismic distribution<br />
alone, as the two patterns are notably different.<br />
Hauck, C. and Vonder Muhll, D. <strong>20</strong>03. Evaluation of geophysical<br />
techniques for application in mountain permafrost studies. Zeitschrift<br />
fur Geomorphologie, Suppl., special issue: geophysics in<br />
geomorphology (in press).<br />
Kneisel, C., Hauck, C. and Vonder Muhll, D. <strong>20</strong>00. Permafrost below<br />
the timberline confirmed and characterized by geo-electric resistivity<br />
measurements, Bever Valley, eastern Swiss Alps. Permafrost<br />
and Periglacial Processes 11: 295-304.<br />
Mamdani, E.H. and Assilian, S. 1975. An experiment in linguistic<br />
synthesis with a fuzzy controller. International Journal of Man-<br />
Machine Studies, 7 (I): 1-13.<br />
Vonder Muhll, D., Hauck, C., Gubler, H., McDonald, R. and Russill, N.<br />
<strong>20</strong>01. New geophysical methods of investigating the nature and<br />
distribution of mountain permafrost. Permafrost and Periglacial<br />
Processes, 12 (1): 27-38.<br />
4.8<br />
I<br />
4ti<br />
4.4<br />
4.2<br />
4<br />
38<br />
3.6<br />
34<br />
Figure 2. Resistivity (left, in loglo(Qm)) and seismic velocity (center, in<br />
kmls) of the synthetic data set, and furzy model output (right, possibility<br />
of ground ice).<br />
As can be seen from Figure 2, the fuzzy model can be<br />
applied to detect areas with high possibility for ground-ice<br />
occurrence from joint resistivity-velocity data sets. In an<br />
initial evaluation study, the model was applied to several<br />
field data sets from anomalously cold low altitude sites in<br />
the midlands of Central Europe and the Alps, where permafrost<br />
is sporadic or absent in general. However, ground<br />
ice may be present in isolated patches with specific microclimatic<br />
conditions (Kneisel et al. <strong>20</strong>00, Gude et al. <strong>20</strong>03).<br />
First results show anomalies with an enhanced possibility<br />
for ground-ice occurrence in comparison to the neighbouring<br />
areas. Even though neither the probability nor the<br />
actual ice content can be determined by the model, the<br />
possibility for ground ice can be quantitatively compared<br />
between different the field sites.<br />
As the fuzzy inference system and the governing decision<br />
rules can easily be adapted to different types of input<br />
data, the applicability of the fuzzy logic approach may be<br />
wide.<br />
REFERENCES<br />
Gude, M., Dietrich, S., Mausbacher, R., Hauck, C., Molenda, R., Ruricka,<br />
V. and Zacharda, M. <strong>20</strong>03. Permafrost conditions in nonalpine<br />
scree slopes in Central Europe. 8th Int. Conf. On Permafrost,<br />
<strong>Zurich</strong>, <strong>20</strong>03, in press.<br />
58
8th International Conference on Permafrost Extended Abstracts an Current <strong>Research</strong> and Newly Available Information<br />
The spatial extent and characteristics of block fields in Alpine areas:<br />
evaluation of aerial photography, LIDAR and SAR as data sources.<br />
S. Heiner, S. Gruber and *E. Meier<br />
Glaciology and Geomorphodynamics Group, Department of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
*Remote Sensing Laboratories, Department of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
INTRODUCTION<br />
The coarse surface layer in block fields operates as an insulating<br />
layer between the surface climate and the ground<br />
below. Often it acts as a thermal filter, insulating against<br />
the effect of surface warming but readily transmitting the<br />
effect of surface cooling (Harris and Pedersen 1998).<br />
Therefore, knowledge about the extent of coarse surfaces<br />
is important for the successful spatial application and further<br />
development of energy balance models (Stocker-<br />
Mittaz et al. <strong>20</strong>02) or statistical models (e.g. Cruber and<br />
Hoelzle <strong>20</strong>01).<br />
Block fields can be mapped relatively accurately based<br />
on the manual interpretation of aerial photography because<br />
of the textural information and terrain position. This<br />
is a labor intensive process depending on the skills of the<br />
operator. This study uses high resolution aerial photography<br />
and a GIS (Geographic Information System) to<br />
automatically map the spatial extent and characteristics of<br />
block fields. The test site is the Corvatsch area in the<br />
south-eastern part of the Swiss Alps.<br />
METHOD AND DATA<br />
In aerial photographs, block fields are characterized by the<br />
frequency of change between dark and bright areas<br />
caused by the shadows of big blocks. Other areas are<br />
typically much more homogeneous. This effect can be<br />
used to separate blocky areas from areas with fewer<br />
blocks by using statistical methods.<br />
On 17 September <strong>20</strong>02, aerial photographs of the test<br />
area have been taken by the Federal Office of Topography<br />
of <strong>Switzerland</strong>. The ground resolution of these pictures<br />
is about 6 cm.<br />
Two more methods for the remote sensing of surface<br />
roughness appear feasible and are currently under investigation:<br />
(I) a SAR (Synthetic Aperture Radar) image of<br />
JERS (Japanese Earth Resources Satellite) is currently<br />
processed and (2) a flight with a LIDAR (Light Detection<br />
And Ranging, laser altimetry) sensor on board will take<br />
place in the test area this summer. LIDAR data from a<br />
different campaign is already available for testing the<br />
methods.<br />
The correctness of the interpretation and validation of<br />
the extracted coarsenesses is provided by transects obtained<br />
from terrestrial surveying.<br />
PROCESSING AND RESULTS<br />
The aerial photographs are orthorectified and imported<br />
into the GIs. Focal functions are applied on this grid.<br />
These are functions which calculate a new value for a cell<br />
using the value of this cell and its adjacent cells. Three<br />
new grids are calculated: one contains the maximum of<br />
two adjacent cells, and the second the mean of two adjacent<br />
cells. Then the difference between these two grids is<br />
calculated and the sum of ten adjacent cells is written into<br />
the central cell. As a result, areas with a high frequency of<br />
change between dark and bright cells (block fields) will<br />
show high values, whereas more homogeneous areas will<br />
be attributed lower values. Now the grid will be reclassified<br />
into the different classes of surface roughness. The last<br />
step is the resampling of the pixels to the desired resolution<br />
for permafrost modeling.<br />
First results are satisfactory as shown in Figure I. The<br />
block fields could be delineated and the method clearly<br />
shows the potential to distinguish between several classes<br />
of surface roughness. Problems remain in shadow areas.<br />
CONCLUSION AND OUTLOOK<br />
The result presented above clearly shows the potential of<br />
the method for the automatic mapping of block fields using<br />
standard GIS software packages. Problems remain to be<br />
solved in the shadow areas. The availability of this data<br />
will improve permafrost modeling.<br />
The use of JERS SAR and LIDAR data will provide an<br />
evaluation using two more data sources. SAR will be inexpensive<br />
and provide large area coverage but only have<br />
a low spatial resolution and only be able to distinguish a<br />
limited number of roughness classes in alpine terrain. LI-<br />
59
DAR will permit a high spatial resolution and accurate<br />
characterization and quantification of surface structure but<br />
requires significantly more cost and effort.<br />
First attempts to use wavelets in the detection of surface<br />
roughness have shown interesting perspectives and<br />
applications of this method to aerial photographs and LI-<br />
DAR data are underway.<br />
Heiner, S. et ai.: The spatial extent and characteristics of block fields,,,<br />
Figure 1. The picture shows the area around the rockglacier Murtel.<br />
Four classes of surface roughness are determined: highest values are<br />
found on the coarse surface of the rockglacier, whereas the homogenous<br />
surface of the skislope shows lowest values.<br />
REFERENCES<br />
Gruber, S. and Hoelzle, M. <strong>20</strong>01. Statistical modelling of mountain<br />
permafrost distribution - local calibration and incorporation of remotely<br />
sensed data. Permafrost and Periglacial Processes 12 (1):<br />
69-77.<br />
Harris, S. and Pedersen, D.E. 1998. Thermal regimes beneath coarse<br />
blocky materials. Permafrost and Periglacial Processes 9: 107-<br />
1<strong>20</strong>.<br />
Stocker-Mittaz, C., Hoelzle, M. and Haeberli, W. <strong>20</strong>02. Modelling alpine<br />
permafrost distribution based on energy-balance data: a first<br />
step. Permafrost and Periglacial Processes 13(4): 271-282.<br />
60
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available l ~ f ~ r ~ ~ ~<br />
Thermal regime of coarse debris layers in the Ritigraben catchment, Matter<br />
Valley, Swiss Alps<br />
T- Herz, L. King and *H. Gubler<br />
lnsfifut fur Geographie der JLU, Giessen, Germany<br />
*ALPuG, Davos, <strong>Switzerland</strong><br />
INTRODUCTION<br />
The objective of the current study is to investigate the<br />
microclimate within coarse debris layers in high mountain<br />
environments. To quantify the heat exchange between the<br />
local atmosphere and the near-surface ground, special<br />
attention is focused on thermal conditions. Further-more,<br />
their influence on the distribution pattern of discontinuous<br />
mountain permafrost is of specific interest. Based on the<br />
research concept and site description given in Herz et al.<br />
(in press), this poster presents and discusses the initial<br />
data gathered from the measurement equipment<br />
described below.<br />
RESEARCH AREA<br />
The data was recorded in close proximity to the meteorological<br />
station erected in the Ritigraben block slope<br />
(46"l I 'N, 7'51 'E, 2615 m a.s.1.). Block sizes in this area<br />
range from 0.5 up to several cubic meters. For comparison<br />
purposes, the finer-grained substrate types known<br />
as "ski run" and "soil beneath alpine grassland" were also<br />
incorporated in the measurement programme.<br />
MEASURING SETUP<br />
Local climatic conditions are recorded at a meteorological<br />
station. The ground thermal regime is registered in a 30 m<br />
deep borehole (cf. Hen et al. in press).<br />
To measure rock and air temperatures within the block<br />
layer, two types of temperature sensors were constructed.<br />
In both cases the sensor element consists of a high-precision<br />
platinum thin film thermo-meter Pt 1000 113DINB<br />
(LxWxH = 10~2~0.<strong>25</strong> mm). The sensors were connected<br />
to a 5-channel-mini-logger (Type Wickenhaeuser<br />
TL-LOG5) by a half-bridge circuit. The technical components<br />
of the logger and the use of a high-precision reference<br />
resistor in the circuit guarantee excellent long-term<br />
stability of the temperature measurements with a theoretical<br />
resolution of 0.00012"C. A four-point calibration<br />
was done in an ice-water bath for 0°C or a circulated<br />
water bath controlled by an analogue calibration<br />
thermometer (resolution 0.01 "C) for IO,<strong>20</strong> and 30°C,<br />
respectively. The calibration yielded outstanding results,<br />
so that the accuracy of temperature measurements can be<br />
indicated as within at least 0.02"C.<br />
Rock and soil temperature sensor elements were<br />
sealed in closed high-grade steel tubes with a wall<br />
thickness of 0.2 mm. Air temperature sensor elements<br />
were fixed in open steel tubes in such a way, so that the<br />
tip of the Pt I000 sticks out of the tube and is in direct<br />
contact with the surrounding air.<br />
Block layer rock and air temperatures are measured in<br />
a profile of four consecutive measurement points. The<br />
profile follows the drainage way of the local micro relief.<br />
Each measurement point consists of two loggers with five<br />
rock and air temperature sensors installed at different<br />
depths. Four additional measurement points form two<br />
cross-sections to the main profile to also include other<br />
micro topographical positions in the block slope.<br />
Surface and soil temperatures are also recorded in the<br />
substrate types, "ski run" and "soil".<br />
INITIAL RESULTS<br />
Hourly meteorological and daily borehole temperature<br />
data have been available since April <strong>20</strong>02. Microclimatological<br />
data exist so far for the nine-week period<br />
from August 15th to October 18th <strong>20</strong>02. Measurement<br />
intervals were set to 5 minutes for air temperature and 15<br />
minutes for rock and soil temperature sensors during this<br />
period. The borehole temperature data presented in<br />
Figure I display a marked asymmetry. The heat input<br />
during summer penetrates down to only 3.5 m in spite of<br />
extreme temperature gradients of up to 10"Clm. A sharp<br />
bend in the maximum curve occurs at that depth. The<br />
maximum temperature remains very close to 0°C during<br />
the whole summer between 3.5 and 5 m depth.<br />
Penetration of surface cooling during winter reaches down<br />
to 14 m depth, tantamount to the depth of zero annual<br />
amplitude. These differences in ground thermal properties<br />
61
etween "summer" and "winter" conditions can only be<br />
explained by phase change processes at a permafrost<br />
table situated between 3.5 and 5 meters depth during<br />
summer <strong>20</strong>02.<br />
Temperature [$I<br />
-10 -5 0 5 10 15 <strong>20</strong> <strong>25</strong> 30<br />
0<br />
Herr, T. et 81.: Thermal regime of coarse debris.. .<br />
response on a surface temperature decrease, which is not<br />
the case for the soil.<br />
. ". . - . . . . " . .. . . . ." .<br />
"-2 cm (block layer)<br />
. . .". ".. ..<br />
. -. . . " .<br />
-- 140 cm @lock layer)<br />
"-150 cm (soil)<br />
2<br />
4<br />
6<br />
8<br />
10<br />
12<br />
E 14<br />
5<br />
16<br />
[Zv&num temperature ~<br />
I ' ,<br />
Maximum temperature1<br />
"Mean temperature<br />
-<br />
Figure 2. Comparison of block layer and soil temperatures<br />
from September 1st to October 18th, <strong>20</strong>02.<br />
18<br />
<strong>20</strong><br />
22<br />
24<br />
26<br />
28<br />
30<br />
Figure 1. Ground temperature envelope, Ritigraben block slope, 261 5<br />
m a.s.1. (April <strong>20</strong>02 -January <strong>20</strong>03).<br />
The resulting thermal offset in ground temperatures<br />
according to Goodrich (1978) is indicated by a mean<br />
temperature of -0.84"C at 3.5 m depth, which is 15°C<br />
colder than the mean temperature at 0-1 m depth. The<br />
permafrost body can be characterized as "warm" with<br />
mean temperatures of -0.37"C at depth of ZAA and<br />
-0.61"C at 30 m depth. It is remarkable, that absolute<br />
minimum temperatures were measured at 1.5 m depth,<br />
representing the bottom side of the uppermost block of the<br />
surface layer. The bottom side is obviously influenced by<br />
advective air circulation between the block layer void<br />
system and the near-ground atmosphere also under<br />
winter conditions, while its surface is protected from heat<br />
loss by the formation of a snow cover.<br />
Figure 2 compares rock temperatures in the block layer<br />
with temperatures measured in an adjacent soil profile.<br />
The different behaviour of the two deeper curves is of<br />
special interest in this context. Rock temperatures at 140<br />
cm depth still show a daily course, which is not detectable<br />
at a comparable depth in the soil. Altogether the temperature<br />
level in the block layer is lower than in the soil in spite<br />
of higher surface temperature values. Rock temperatures<br />
at 140 cm depth clearly correlate with the daily minima of<br />
the surface temperature course and show an immediate<br />
CONCLUSIONS AND OUTLOOK<br />
The data presented indicate that non-conductive heat<br />
transfer processes play an important role in the thermal<br />
regime of block layers. Advection of water and air in the<br />
void system between blocks is believed to be the crucial<br />
phenomenon, which is investigated in this study.<br />
Furthermore, in October <strong>20</strong>02 five additional boreholes<br />
were drilled in the lower part of the Ritigraben block slope<br />
to investigate the rheological conditions in the catchment<br />
of the torrent.<br />
REFERENCES<br />
Goodrich, L.W. 1978. Some results of a numerical study of ground<br />
thermal regimes. In 3rd International Conference on permafrost,<br />
<strong>Proceedings</strong>, Edmonton, 10-13 <strong>July</strong>: 29-34.<br />
Herz, T., King, L. and Gubler, H. in press. Microclimate within coarse<br />
debris of talus slopes in the alpine periglacial belt and its effects<br />
on permafrost. In 8th International Conference on Permafrost,<br />
<strong>Proceedings</strong>, <strong>Zurich</strong>, 21-<strong>25</strong> <strong>July</strong> <strong>20</strong>03.<br />
62
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Topographical properties of active and inactive patterned grounds in<br />
Finnish Lapland<br />
J. wort and *M. Seppala<br />
Department of Geography, University of Helsinki, Finland<br />
*Physical Geography Laboratory, University of Helsinki, Finland<br />
Periglacial patterned grounds are quite widespread features in<br />
Northern Finnish Lapland, especially above timberline in the<br />
zone of discontinuous permafrost. The most common types<br />
are sorted and non-sorted nets. Activity of these landforms<br />
varies from one area to another and between different types of<br />
patterned grounds. Active forms seem to be mainly located in<br />
the valleys, Respectively, inactive patterned grounds are<br />
more common on the slopes and on top of the fells (Fig. I). In<br />
general, climate is a crucial environmental determinant of frost<br />
activity. However, when we study regions at large scale the<br />
importance of local factors such as soil, hydrology, topography,<br />
vegetation etc., becomes apparent.<br />
The objectiveis to present some results of a research<br />
where active and inactive patterned grounds were studied<br />
using Geographical Information System (GIS) and statistical<br />
methods. The main aim was to compare topographical pmperties<br />
of active and inactive regions to see if there are any<br />
differences between these areas. Topographical parameters<br />
calculated from a digital elevation model (DEM) were used<br />
as explanatory variables in univariate analysis. The DEM<br />
with <strong>20</strong> m grid size was created from contours of topographic<br />
maps using Arcllnfo 8.1. SPSS 9.0 software package was<br />
usedinstatisticalanalysis.<br />
Figure 1. Distribution of active and inactive patterned grounds in a part (4.8 x 10 km) of the study area (10 x <strong>20</strong> km) at Paistunturit fell region<br />
in Finnish Lapland (Contours 0 National Land Survey of Finland, Licence number 491MYY103).<br />
63
Table 1. Results of an univariate analysis where active and inactive patterned ground squares were compared with topographical variables<br />
calculated from DEM. Total number of the grid squares in the analysis is 745. Statistical significance is tested with the Mann-Whitney U-test.<br />
Topographical Patterned ground Statistical<br />
variables<br />
(PI<br />
Active (n 421) Inactive (n = 324)<br />
mean f std<br />
mean f std<br />
Mean altitude (m a.s.1.) 392 f 40 454 f 49
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available ~~~~r~~~~~~<br />
Influence of human activities and climatic change on permafrost at<br />
construction sites in Zermatt, Swiss Alps<br />
R. Hot 1. King, T. Hen and *S. Gruber<br />
Institute for Geography, Justus Liebig University, Giessen, Germany<br />
*Glaciology and Geomorphodynamics Group, Department of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
GEOGRAPHICAL BACKGROUND<br />
PERMAFROST CONDITION<br />
AND<br />
(c) other related structures such as masts, tunnels, elevators,<br />
shelters for vehicles, workshops, etc.<br />
(d) subsurface water pipes (for drinking water, artificial<br />
snowing of ski-runs), sewage, communication and electricity<br />
Zermatt is a tourist resort in the Alps with a total of 14,000<br />
hotel beds and more than one million oficial overnight stays lines.<br />
per year. The necessary infrastructure consists of installations Engineering geologists as well as those responsible perthat<br />
often reach into permafrost areas, from the sporadic zone<br />
at about 2600 m a.s.1. up to the continuous permafrost zone<br />
sons for this infrasttucture have become increasingly interested<br />
in the distribution and the characteristics of permafrost in<br />
above 3400 m a.s.1. The unglaciated permafrost area has a the Zermatt area, as there have been problems due to perma<br />
large vertical extension due to the surrounding high mountain frost degradation.<br />
ranges that reach above 4000 m a.s.l., resulting<br />
a dry and The poster gives an inventory of the existing structures on<br />
sunny climate and a very high glacier equilibrium line.<br />
The infrastructure erected in the permafrost areas consists<br />
of<br />
probable and proven permafrost sites and describes some<br />
problems encountered during the last <strong>25</strong> years, with original<br />
new temperature data.<br />
(a) hotels, restaurants and mountain huts,<br />
(b) station buildings of railways, funiculars and ski lifts,<br />
0 50m<br />
Figure 1. Map of Kleinmatterhorn mountain peak with tourist facilities (IO meter contour lines).<br />
x temperature logger<br />
65
CONSTRUCTION SITES<br />
Kulrnhotel and Gornergrat Funicular<br />
Gornergrat can be reached by a 9.340 km long railway line<br />
constructed between 1896 and 1898. The uppermost 500<br />
meters were added in the year I909 and the Kulmhotel Gornergrat<br />
was opened in 1910. The railway track and the hotel<br />
were mainly used in summer until the early Fifties. Winter<br />
skiing became more and more attractive in the Sixties when<br />
the hotel started to open also in winter. In the year 1985 two<br />
new astronomic observatories were added to the northern<br />
and southern towers and this caused a substantial additional<br />
load on the subsurface. Moreover, apartments for scientists<br />
were heated in the basement.<br />
Soon after this, the northern tower that was erected on<br />
frozen morainic material started to settle due to permafrost<br />
degradation. The subsidence between tower and hotel was<br />
controlled with geodetic measurements and strain gauges and<br />
concrete were injected into the foundation in order to stop the<br />
differential settlement. Today, ground temperature measurements<br />
show that the ground below the tower is probably not<br />
frozen for a depth of about 10 meters (the active layer at un-<br />
disturbed sites is less than 2.5 meters). This must be attrib<br />
uted to the heating of the basement for almost <strong>20</strong> years. Actually,<br />
subsidence seems to have stopped.<br />
n of, R, et ai.: Human activities, climatic change, permafrost, Zermatt ...<br />
In view of the considerable temperature rise in the bedrock,<br />
temperatures havebeenmonitoredsince1998at ten<br />
different places in order to takecountermeasures if necessary.<br />
2.3 Further construction sites on permafrost<br />
There are many more facilities erected on the permafrost (Table<br />
l). Climatic change may further contribute to a rise in the<br />
lower permafrost limit. It should be recalled that an actual<br />
MAAT of about -1°C indicates patchy permafrost occurrences,<br />
and at less than -5T, continuous permafrost may be<br />
expected.<br />
Table 1. Installations on permafrost in the<br />
tudes and expected MAAT.<br />
Zermatt area with alti-<br />
Name<br />
Altitude (m a.s.1.) MAAT (“C)<br />
Kleinmatterhorn<br />
(station) 38<strong>20</strong> -7,2<br />
(elevator summit) 3883 -7,6<br />
Testa Grigia (station) 3479 -53<br />
Stockhorn (station) 3407 -4,9<br />
Hohtalli (station) 3286 -4,2<br />
Matterhorn Refuge 3260 -4,O<br />
Rote Nase (station) 3<strong>25</strong>0 -4,O<br />
Kulmhotel Gornergart 31 35 -3,3<br />
Rothorn (station) 31 03 -3,l<br />
Gandegg Refuge 3029 -2,7<br />
Trockener Steg (station) 2939 -2,2<br />
Gornergrat rack railroad (top) 3090 -3,O<br />
CONCLUSIONS<br />
Figure 2. Kleinmatterhorn mountain top with the location of temperature<br />
loggers. The position of the cross sections A, B and C are<br />
shown in Figure 1.<br />
Degradationofpermafrostdue to climatic change may be<br />
come a serious threat to installations at high mountain tourist<br />
resorts. This is especially dangerous when the facilities are<br />
not properly maintained. Therefore, relevant information must<br />
be improved to the managers of these installations and to the<br />
local authorities by permafrost scientists working in these areas.<br />
Kleinmatterhorn Funicular<br />
This funicular travels up to 38<strong>20</strong> m a.s.1. and were constructed<br />
in 1981, It arrives at a tunnel cut in the northern wall<br />
of the mountain top (Fig. 1). At its southern exit the ski tun<br />
starts down to Zermatt. Bedrock temperatures as low as<br />
-12°C were reported during the construction (Keusen and<br />
Haeberli 1983). The same temperature was measured in the<br />
near-surface rock material of the northem rockwall below the<br />
mountain top in a diploma study. The MAAT measured here<br />
was -8°C in the years 1998199.<br />
Today, bedrock temperatures have risen to -3 to -2°C at<br />
several localities (cf. Fig. 2). This is due to heating and to the<br />
heat brought into the tunnel by the more than 490,000 visitors<br />
per year. Additional heat is created by almost 70,000 elevator<br />
movements per year, transporting tourists to the mountain top.<br />
In summer 1997, meltwater created problems when re<br />
freezing in the elevator shaft (King et al. 1998).<br />
REFERENCES<br />
Keusen H.R. and W. Haeberli, W. 1983. Site investigation and<br />
foundation design aspects of cable car construction in Alpine<br />
permafrost at the “Chli Matterhorn”, Wallis, Swiss Alps. In 4Ih International<br />
Conference on Permafrost, <strong>Proceedings</strong>, Fairbanks,<br />
17-22 JuIv: 601-605.
8th International Conference an Permafrost Extended Abstracts an Current <strong>Research</strong> and Newly Available ~ n f ~ r ~ ~ ~ i o<br />
Spatial conditions of frozen ground and soil moisture at the southern<br />
boundary of discontinuous permafrost in Mongolia<br />
M. Ishikawa, Y, Zhang, T. Kadota, T. Ohata and N. Sharkhuu<br />
Frontier Observational <strong>Research</strong> System for Global Change, Yokohama 236-0001, Japan<br />
*Institute of Geography, Mongolian Academy of Sciences, Ulaanbaatar 2106<strong>20</strong>, Mongolia<br />
INTRODUCTION<br />
As a consequence of global warming, mean annual air temperatures<br />
are predicted to increasesignificantlyinthehigh<br />
latitudes, leading to changes in the distribution and geometries<br />
of permafrost on a decadal scale. Degradation and thawing of<br />
permafrost is expected to occur, especially atthesouthern<br />
boundaries of discontinuous permafrost, resulting in significant<br />
changes in the components of land-surface hydrology such<br />
as surface runoff, groundwater flow, soil moistures and<br />
evapotranspiration as well as those of energy balance.<br />
The Frontier Observational <strong>Research</strong> System for Global<br />
Change has promoted a research program entitled “Observational<br />
study on the water cycle in the periphery of Eurasian<br />
snow cover/frozen ground region in relation to climate forma<br />
tion”. The critical components of this project are to clarify and<br />
to define the hydrological roles of frozen ground with respect<br />
to land-surface hydrology. This year we focused on the spatial<br />
variations of ground thermal and moisture conditions in<br />
Mongolia at the southern boundary of Eurasian discontinuous<br />
permafrost. On the basis of the results obtained, this paper<br />
discusses the local distribution and thermal condition of permafrost,<br />
and its hydrological roles.<br />
REGIONAL SETTING OF STUDY AREA<br />
The Tuul river basin in the Khentei Mountains, northeast d<br />
Ulaanbaatar was selected as an area for intense observation.<br />
The topography is characterized by a significant relief with altitude<br />
ranging from 1300 to <strong>25</strong>00 m A.S.L. Mean annual air<br />
temperature (MAAT) averaged between 1992 and <strong>20</strong>00 was<br />
-3.5 ‘C at the Terelj meteorological station (15<strong>20</strong> m A.S.L).<br />
The <strong>20</strong> years MAATs averaged at Zuunmod (1529 m A.S.L.)<br />
and at the Ulaanbaatar stations (1300 rn A.S.L.) were close<br />
to -1.0 ‘C. Recent long-term air temperature observations indicate<br />
significant warming. Originally negative MAATs have<br />
sometimes become positive values after the late 1990s.<br />
Mean annual total precipitation<br />
Ulaanbaatar station was 300<br />
mm between 1980 and <strong>20</strong>00, and approximately 80 % of the<br />
precipitation occurs between June and August.<br />
In spite of such dry climatic conditions, dense forests are<br />
extensively developed mainly on the northern slopes, while<br />
pastures dominate on the southern slopes and on the flat<br />
plains.<br />
OBSERVATIONS<br />
Observations include measurements of ground temperatures<br />
at shallow and greater depths, of soil moisture and of hydraulic<br />
conductivity at two locations; in Shijir valley, west of the<br />
Terelj station and at the Nalaikh site. In the Shijir basin, hnsect<br />
measurements were carried out on steep and gentle<br />
northern forest slopes (NFSs) and on southern pasture slopes<br />
(SPSs) in the rainy season of <strong>July</strong> and during the dry season<br />
of October and November <strong>20</strong>02. At the Nalaikh site, we<br />
drilled a borehole reaching a depth of 30 m in early October.<br />
The drill site is on the wide flat pasture plain (FPP). A ground<br />
temperature profile was measured on the 1st November.<br />
Core samples were taken at <strong>20</strong> cm depth intervals and used<br />
for estimating soil water contents.<br />
RESULTS<br />
Shjiir valley (Figure 1)<br />
Steep NFSs (sites A, B, C, Q and R) havelarge angular<br />
gravels overlain by thick organic materials composed of hu-<br />
mus and moss, while gentle NFSs (sites E and F) dominate<br />
silt materials. Soil textures beneath SPSs were characterized<br />
by sand-gravels overlain by thin humus. The highest hydraulic<br />
conductivity (I00 mld) was found at steep NFS (site A).<br />
The values on other slopes ranged from 0.1 to IO mld. Ice<br />
lenses were visible only beneath steep NFSs only. Higher<br />
ground water tables were found at gentle NFSs.<br />
Ground temperatures at 1.5 m depth were low on steep<br />
NFSs (reaching 0 ‘C at sites A, B, C, Q and R), but higher<br />
(ranging from 5 to 6 OC at sites I, Nand 0) on SPS. Thermal<br />
observations at greater depths in November indicated the OG<br />
currence of permafrost below a depth of 2.4 m on NFS (site<br />
E). On the other hand, ground temperatures at SPSs were<br />
about 5 ‘C at the same depths, suggesting that permafrost is<br />
absent.<br />
67
High moisture contents, reaching 40 and 60 % at some<br />
depths, occurred beneath gentle NFS (sites E, F and S). At<br />
these sites, ground water tables were found at shallower<br />
depths. On the other hand, moisture contents were less than<br />
<strong>20</strong> % on steep NFSs (A, B and C) and SPSs (/and 0).<br />
Nalaikh site (Figure 2)<br />
At the Nalaikh site, freezing reached a depth of 0.9 m on b<br />
1st of November. The highest temperature (3.7 'C) occurred<br />
at 3 m depth. Below this depth, ground temperature de<br />
creaseduntiltheuppermostportion of frozen grounds was<br />
reached. Subzero temperatures (0 to -0.2 'C) occurred be<br />
tween 6 and 12 m depth, indicating permafrost conditions. The<br />
temperatures of sub-permafrost layers were slightly above 0<br />
OC.<br />
Significant differences were found in the water contents d<br />
the active layer, the permafrost and sub-permafrost layers,<br />
respectively. While water contents were less than 10 % in<br />
the active layer, those in the upper portion of permafrost were<br />
high, varying from 15 Yo to 27 %. The water contents in the<br />
lower portions of permafrost were from IOto <strong>20</strong> % and similar<br />
to those of sub-permafrost layers.<br />
SPATIAL THERMAL CONDITIONS OF PERMAFROST<br />
Generally, permafrost underlies the NFSs where the incoming<br />
local energy fluxes on the surface are reduced due b<br />
slope orientation, forest shading and thermal properties d<br />
substrata. Cold permafrost with a thinner active layer occurs<br />
beneath NFSs.<br />
Extremely warm permafrost with thick active layers sporadically<br />
occurs beneath FPPs. According to the increasing<br />
rate of ground temperature change estimated by Sharkhuu<br />
(<strong>20</strong>03), the permafrost at Nalaikh site could completely thaw<br />
during the coming <strong>20</strong> years.<br />
lshikawa, M, et 31.: Spatial conditions of frozen ground and sol ...<br />
ables water to exchange between the bottom of the active<br />
layer and the uppermost part of the permafrost. Such permeable<br />
permafrost would not keep the overlaying active layer<br />
wet. As suggested in the unevenly distributed forest, this<br />
process may affect the land-surface hydrology even within<br />
the small areas and needs further studies with due consideration<br />
of the differences in permeability between contrasting<br />
sites with cold and warm permafrost.<br />
Figure 1. Spatial pattern of thermal and moisture conditions in the:<br />
Shijir valley in mid-<strong>July</strong>, <strong>20</strong>02: (a) Topography and location of each<br />
site (b) Ground temperature profiles (c) Volumetric water contents,<br />
depths of groundwater level and ground ice.<br />
DISCUSSION<br />
The influence of permafrost as an impermeable layer can be<br />
well delineated and observed in comparisons with the distribution<br />
pattem of soil moisture in the Shijir valley. Water from<br />
precipitations was kept in active layers, leading to high<br />
moisture contents and groundwater levels beneath gentle<br />
NFSs. The differences in hydraulic conductivity and topography<br />
between steep and gentle NFSs enable subsurface water<br />
to flow downward laterally above the permafrost table and<br />
to stay beneath gentle NFSs, resulting in high moisture contents<br />
only beneath gentle NFSs (Fig. 1). Dry subsurface<br />
conditionson SPSs are likely to be dominated by vertical<br />
water movements, because of the absence of impermeable<br />
layers such as permafrost.<br />
Vertical water movements also probably occur at the<br />
Nalaikh site underlain by extremely warm permafrost with<br />
dry active layers. Since the amount of unfrozen water in b<br />
zen soils increases as temperature increases to zero (e.g.<br />
Burt and Williams 1976), large amounts of unfrozen water are<br />
contained within the permafrost. This condition probably en-<br />
68<br />
Figure 2. Thermal (1st November) and moisture<br />
profile at Nalaikh site.<br />
REFERENCES<br />
(early October)<br />
Burt, T. P. and Williams, P.J. 1976. Hydraulic conductivity in frozen<br />
soils. Earth Surface Processes, 1: 349-360.<br />
Sharkhuu, N. in press. Recent changes in the permafrost of Mongolia.<br />
In: <strong>Proceedings</strong> of the 8th International Conference on Permafrost.<br />
<strong>Zurich</strong>, <strong>Switzerland</strong>.
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available ~ f l ~ ~ ~<br />
Radio wave method for the determination of the electrical properties of<br />
permafrost in crosshole space<br />
V.A. Isfrafov, S.O. Perekalin, A.O. Kuchmin and *A. D. Frolov<br />
Radionda Ltd, Moscow, Russia<br />
*Scientific Council on Earth Ctyology, Russian Academy of Sciences, Moscow, Russia<br />
Geocryology studies are important during the construction<br />
and exploitation of engineering objects in permafrost areas.<br />
Investigations of inter-borehole space by the radio<br />
wave geo-introscopy (RWGI) method, whereby resistivity<br />
and dielectric permittivity are simultaneously defined,<br />
make it possible to ensure higher reliability when determining<br />
the forms and disposition of the geological inhomogenities<br />
within the volume under study (Borisov et al.<br />
1993). RWGl measurements could be a reliable instrument<br />
for the delineation of the frozen - thawed sediment<br />
boundaries. This suggestion is made, first of all, because<br />
of the high dependence of formation resistivity and dielectric<br />
permittivity on the aggregate status of porous water.<br />
The first tests in this domain were carried out in 1940<br />
near lgarka (Eastern Siberia). These experiments have<br />
shown (Petrovsky and Dostovalov 1947) that the propagation<br />
of radio waves (up to frequency - 6.5 MHz) is quite<br />
possible through about 10 m of permafrost. Since this time<br />
there have been many tests using various radio wave<br />
techniques to study frozen formations. It needs to be<br />
noted that until recent years there was little evidence of a<br />
concrete technology (special equipment, procedures for<br />
field measurements, data processing, mode of result<br />
presentation etc.) to make such studies.<br />
The physical-geological basis of the RWGl method is<br />
the relationship between the intensity of radio energy absorption<br />
along the wave path and the electrical characteristics<br />
of the soils. The coefficient of radio wave absorption<br />
(k") depends on the electromagnetic field frequency and<br />
the electrical properties of the medium:<br />
I<br />
o =2II f, f is the operating frequency, p and E are,<br />
respectively, the absolute magnetic and dielectric<br />
permittivities, p is the resistivity of a medium.<br />
In soils with a resistivity 400 am, the radio wave absorption<br />
is practically not dependent on the dielectric permittivity<br />
(the medium is a quasi-conductor).<br />
In frozen soils with high resistivity (p -hundreds to thousands<br />
Qm) and E ~30, radio wave absorption at<br />
frequencies of 30 MHz and higher depends considerably<br />
on permittivity (the medium is quasi-dielectric). When<br />
measurements are performed in an inhomogeneous medium,<br />
k" is an integral value that summarizes all the local<br />
changes in the medium absorbing properties along the<br />
wave path, or more correctly, within the most essential<br />
space for radio wave propagation (the first Fresnel zone).<br />
Having performed measurements at two fixed frequencies,<br />
it is possible to calculate the effective magnitudes of electric<br />
resistivity pe and dielectric permittivity &e of soils within<br />
the studied intervals.<br />
A special digital borehole radio equipment of the<br />
RWGI-2 series was designed and manufactured by "Radionda<br />
LTD" for measurements of the intensity of the electromagnetic<br />
field emitted using an electric (or magnetic)<br />
dipole antennae at the fixed frequencies of 0.061, 0.156,<br />
0.6<strong>25</strong>, 1.<strong>25</strong>, 2.<strong>25</strong>, 10.0 and 31 .O MHz. Cross-borehole<br />
measurements are usually performed following the "tomographic<br />
survey" technique, where the field is measured at<br />
each position of the transmitter in one borehole, over the<br />
whole working range of receiver displacement in the other<br />
borehole, - "in fan manner" (Fig. 1). To exclude the antenna<br />
effects of the cable, the borehole units are connected<br />
to the cable through dielectric inserts (coupler) with<br />
a fiber-optic channel and an optoelectronic converter. Besides<br />
measuring the electric (or magnetic) field intensity at<br />
the reception point and those of the current in the transmitting<br />
antenna, the equipment design makes it possible<br />
to execute transmitter tuning at each position in the borehole<br />
and to control the emission power. This makes it<br />
possible to operate in boreholes with distances of<br />
hundreds of meters from each other even with a short<br />
antennae. While processing, the measured data is corrected<br />
to exclude any differences in power emission. In<br />
the upper part of Figure I the scheme of RWGI-CB measurements<br />
along 6 sections with 4 boreholes (1-4) is presented<br />
as an example. The contour of the first Fresnel<br />
zone is shown for each section (between any pair of<br />
holes).<br />
To obtain reliable final data, it is essential to secure a<br />
69<br />
sufficient overlapping of these zones. To do this requires a<br />
preset and suitable operating frequency for the given con-
~~~~~~ CIPEI:<br />
UT<br />
ditions for RWGI-CB. All this ensures the most complete<br />
and uniform studies of inter-hole space because every inhomogenity<br />
in the medium is explored at various angles.<br />
~ i ~ wf i ~ - ~ ~<br />
es ~~~~~~~<br />
lstratov, V.A. @t al.: Radio wave method for the determination ...<br />
G%rn.l m k %<br />
d”<br />
each borehole. In is evident that by means of RVGl it is<br />
possible to extrapolate high temperature zones in an interhole<br />
media as areas of low resistivity. The most conductive<br />
zone corresponds to a leakage channel.<br />
The briefness of this paper does not permit the discussion<br />
of the very interesting results of RVGl monitoring and<br />
the measurements of dielectric permittivity obtained at the<br />
same site (Snegirev et al. <strong>20</strong>03).<br />
The modern technology of RWGl measurements and<br />
associated data processing provides geocryological investigations<br />
with an appropriate resolution for engineering<br />
purposes.<br />
REFERENCES<br />
Borisov, B.F., Istratov, V.A. and Lysov, M.G. 1993. Method of radio<br />
wave cross-borehole geointroscopy. Patent of Russia <strong>20</strong>84930.<br />
Petrovskii A.A. and Dostovalov B.N. 1947 First experience of radiowave<br />
propagation in permafrost media. <strong>Proceedings</strong> of the<br />
Permafrost Institute 5: 121-160. Moscow: USSR Academy of Sciences<br />
press.<br />
Snegirev, A.M., Velikin, S.A., Istratov, V.A., Kuchmin, A.O. and Frolov,<br />
A.D. <strong>20</strong>03. Geophysical monitoring in permafrost areas. <strong>Proceedings</strong><br />
of the 8th International Conference on Permafrost, <strong>Zurich</strong>,<br />
<strong>Switzerland</strong>.<br />
Figure 2. Example of the geoelectric section of effective resistivity<br />
(fragment of the 3D map) obtained from the field studies made by the<br />
RWGI-CB technique at the site near diamond pipe “Udachnaya”.<br />
As an example let us consider the investigations performed<br />
in order to delineate subsurface leakage channels<br />
from the Sytykan water reservoir (Yakutia). The crosshole<br />
RWGl measurements were provided using 1.<strong>25</strong> MHz. A<br />
fragment of the geoelectrical map is shown in Figure 2.<br />
The section is orthogonal to the shoreline of the Sytykan<br />
reservoir. The temperature curves are also shown near<br />
70
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Effects of volumetric ice content and strain rates on shear strength ant 1<br />
creep rate under triaxial conditions for frozen soil samples<br />
M. M. Johansen and *l. U. Arenson<br />
Department of Civil Engineering, BYGDTU, Technical University of Denmark, Lyngby, Denmark<br />
*Institute for Geotechnical Engineering, Swiss Federal lnstitute of Technology, <strong>Zurich</strong>, <strong>Switzerland</strong><br />
INTRODUCTION<br />
Rock glaciers are special geomorphological phenomena d<br />
mountain permafrost. In some rock glaciers that have been<br />
investigated recently, a clear shear horizon has been de<br />
tected (Arenson et al. <strong>20</strong>02) in which most deformation takes<br />
place. Within an alpine environment, rock glaciers may have<br />
temperatures close to 0 "C(Vonder Miihll et al. <strong>20</strong>03), which<br />
make them very sensitive to global warming. Even though<br />
rock glaciers in the Swiss Alps have been investigated intensively<br />
since the 1970s (e.g. Barsch 1996), there are still a<br />
lot of unanswered questions. In particular, the loss of strength<br />
of the frozen geomaterial, which might lead to stability prob<br />
lems of steep slopes, is expected as an effect of a warmer<br />
climate and is of major concern in <strong>Switzerland</strong>.<br />
Therefore, a set of triaxial creep and shear tests were<br />
performed to study the effect of the volumetric ice content under<br />
natural stress conditions.<br />
TESTS<br />
The goal for this project was to examine mechanical re<br />
sponse of various ice-solid mixtures from triaxial shear and<br />
creep tests. A programme was set up to test the strength under<br />
different confining pressures and for a range of strain rates<br />
as well as volumetric ice contents. The material was prepared<br />
as closely as possible to the natural conditions found in<br />
alpine rock glaciers (e.g. Arenson et al. <strong>20</strong>02, Vonder Muhll<br />
et al., <strong>20</strong>03) and tested at equivalent temperatures. To obtain<br />
controllable and repeatable tests, the original soil was mixed<br />
in the laboratory. Triaxial shear tests and creep tests were<br />
carried out in a cold room with a fixed temperature at the Institute<br />
for Geotechnical Engineering at the Swiss Federal Institute<br />
of Technology in <strong>Zurich</strong>. Artificially-made samples with<br />
an ice content of 30, 50, 80 and 100% by volume, respectively,<br />
were tested under changing testing parameters. The<br />
confining pressures s3 were 50, I00 and <strong>20</strong>0 kPa and the<br />
strain rates used for the shear tests were 0.<strong>25</strong>, 2.5 and<br />
<strong>25</strong> mmlh. All tests were carried out at a temperature of about<br />
-1 S"C. During a test, the temperature variations were about<br />
+/-0.05 "C.<br />
TEST RESULTS<br />
for<br />
Figure 1 shows the peak and the residual shear strength<br />
samples with different ice contents at three different strain<br />
rates. Figure 2 shows the peak and residual shear strength<br />
for samples with different strain rates at four different ice contents.<br />
The tests indicate that the strength of the frozen soil increases<br />
rapidly at the beginning of the test, probably due b<br />
icestrengthening,after which itreachespeakstrength and<br />
drops to a lower value due to bonds cracking between Ute<br />
soil and the ice. The authors observed no clear trend between<br />
the strength and the confining pressure within the range d<br />
confining pressure tested, especially for ice-rich soils. A trend<br />
was recognized for soils with lower ice contents (30%).<br />
Soils with low volumetric ice content and high confining pressure<br />
showed a behaviour similar to unfrozen soils. Soil<br />
strengthening takes place due to structural hindrance of the<br />
solid particles after an initial slight strengthening of the ice with<br />
increasing solid content. This results in an increase of the<br />
strength during the test with increasing axial strain, which has<br />
also been described by other authors (e.g. Ladanyi 1981,<br />
Ting et al. 1983).<br />
t<br />
I<br />
0- i<br />
0.1 1 10 100<br />
Figure 1. Shear strength (s1-s3) versus strain rate for different ice<br />
contents. Confining pressure $3 = 100 kPa.<br />
71
Johansen, M.M. et al.: Effects of volumetric ice content and strain rates un shear strength ...<br />
E 3000<br />
with four different ice contents. To observe trends better,<br />
s<br />
samples with different ice contents should be prepared and<br />
" <strong>25</strong>00<br />
tested. In addition, the influence of the air in the sample and<br />
0<br />
<strong>20</strong>00<br />
the difference between peak strength and high strain rates<br />
should be examined.<br />
1500<br />
The reason for testing artificially-made samples instead d<br />
1000<br />
original samples extracted from a rock glacier is that it should<br />
be possible to carry out repeatable tests. This is often not<br />
possible when original samples are used, because the composition<br />
of the sample can vary and this not well recognized<br />
0 <strong>20</strong> 40 60 80 100 1<strong>20</strong><br />
until after the test. Furthermore, the recovery of original per-<br />
Volumetric ice content mafrost samples is very expensive due to the costs of drilling<br />
[" +a. .<strong>25</strong>, peak . rn 2.5. peak ~ <strong>25</strong>, peak b0.<strong>25</strong>, res. o 2.5, res. 1<strong>25</strong>, res] and fieldwork. However, it was the intention to prepare and<br />
test the samples under similar conditions found in a real rock<br />
glacier. Further investigations on artificial soil samples may<br />
help in developing a better understanding of rock glacier dynamics.<br />
There is a clear trend that higher strain rates result in<br />
higher shear strengths. In general it is found that the strength<br />
increases with decreasing ice content except for the samples<br />
REFERENCES<br />
with 100% ice, which showed the highest strength (Fig. 1). Arenson, L.U., Hoelzle, M. and Springman, S.M. <strong>20</strong>02. Borehole<br />
The authors assume that this is because a different mecha- deformation measurements and internal structure of some rock<br />
nism is in play. The sample showed a brittle behaviour with a glaciers in <strong>Switzerland</strong>. Permafrost and Periglacial Processes<br />
sudden loss of strength after a peak value was reached. 13: 1 17-135.<br />
Because of limited time, only a few creep tests were car-<br />
Barsch, D. 1996. Rockglaciers: indicators for the present and<br />
former geoecology in high mountain environments.<br />
ried out. These could not be used for proper analysis. Further Springer series in physical en vir0 nmen t 16.<br />
tests should be carried out to examine the creep behaviour d<br />
the frozen soil in more detail.<br />
Figure 2. Shear strength (m-03) versus volumetric ice content at<br />
different strain rates. Confining pressure o3 100 kPa.<br />
CONCLUSIONS<br />
The following conclusions can be drawn from the tests on b<br />
Zen soils presented.<br />
Samples have to be prepared very carefully in order b<br />
achieve repeatable tests. The ends of the samples should be<br />
cut precisely and parallel to each other so that the stress is<br />
equally distributed on the surface. The technique used for<br />
sample preparation has proved to be appropriate, but an air<br />
content of about 4% was measured. To avoid air in the Sample,differentmethods<br />
for thesamplepreparationshould be<br />
considered.<br />
The strength and the mechanisms of frozen soil are very<br />
dependent on the temperature. Lower temperature leads b<br />
higher strength. Small temperature differences in the soil<br />
samples and in the cold room used for tests can explain the<br />
different values for strength found for similartests.On the<br />
other hand, small differences in the volumetric ice content<br />
might also have influenced the strength of the sample.<br />
Due to the time limitations of this diploma project, it has not<br />
been possible to examine everything in depth. In additions,<br />
questions have arisen during the testing of the samples considering<br />
mechanical behaviour of frozen soil which are worth<br />
investigating further. The volumetric change in the samples<br />
should be examined and a better method to measure the volume<br />
of the sample should be developed.<br />
The ice content versus shear strength should be examined<br />
in more detail. The tests were only made on samples<br />
72<br />
Ladanyi, B. 1981. Mechanical behaviour of frozen soils. Mechanics<br />
of Structured Media B: <strong>20</strong>5-245.<br />
Ting, J., Martin, T. and Ladd, C. 1983. Mechanisms of strength for<br />
frozen sand. J. of Geotech. Eng. 109: 1286-1302.<br />
Vonder Muhll, D. Arenson, L.U. and Springman, S.M. <strong>20</strong>03. Temperature<br />
conditions in two Alpine rock glaciers. Proc. 8" Int.<br />
Conference on Permafrost, <strong>Zurich</strong>: this issue. Lisse: Swets &<br />
Zeitlinger.
Evolution of Alpine permafrost under the impact of climatic<br />
fluctuations (the results of numerical modeling)<br />
M. Kasymskaya, *D. Sergueev and N. Romanovskii<br />
lomonosov Moscow State University, Faculty of Geology, Geocryology Department, Leninsky Gory, Russia<br />
*Geophysical Institute, University of Alaska, Fairbanks, Alaska, USA<br />
Alpine permafrost is characterized by considerable<br />
spatial and temporal variability, especially near its southern<br />
boundary. To assess the evolution of alpine permafrost<br />
we used a 2-D computer-based model developed by<br />
G.S. Tipenko. This model takes into account (1) long-term<br />
climatic fluctuations, (2) the geometry of mountain relief,<br />
(3) the thermal properties of rocks and their moisture<br />
content and (4) the character of permafrost altitudinal<br />
zonality.<br />
This model was applied to study permafrost evolution<br />
within a typical mountainous area in Southern Yakutia<br />
(Chul'man Depression) with its extremely continental climate,<br />
flat-topped mountains and a discontinuous distribution<br />
of permafrost. This area has an inverse type of altitudinal<br />
zonality, Le., mean annual temperatures of both air<br />
and rocks rise as the absolute height increases. A profile<br />
across a mountain massif with flat-topped divide (950 m<br />
a.s.1.) and depression (750 m a.s.1.) was selected for<br />
modeling. (Fig.1)<br />
i.0<br />
I I I I I<br />
0 14 <strong>25</strong> 36 63 74<br />
Figure 1. Chart of the model field<br />
The new palaeogeographical scenario developed by<br />
A.V. Gavrilov for the last 1<strong>20</strong> ka was used (Fig.2) This<br />
scenario takes into account the effect of permafrost altitudinal<br />
zonality on the dynamics of mean annual ground<br />
temperatures. It was assumed that ground temperatures<br />
on the tops of plateaus (divides) rise above 0°C in the periods<br />
of maximum climatic warming.<br />
Temperature, "C<br />
-12 -10 -8 -6 -4 -2 0 2 4<br />
1- on the mountain top -2- on the valley bottom<br />
Figure.2. The curves of mean annual ground temperatures in accordance<br />
with the pako- geographic scenario.<br />
1<br />
Numerical modeling showed that in the given type of<br />
mountains, permafrost temperatures in depressions and<br />
river valleys are substantially higher than those on watersheds.<br />
In periods of climatic warming, taliks can develop<br />
on the tops of plateaus. Thus, infiltration of atmospheric<br />
precipitation becomes possible in summer periods. It was<br />
also found that the maximum depth of permafrost is<br />
reached, with some time lag, after the minimum permafrost<br />
temperature has been attained.<br />
The results of this modeling make it possible to estimate<br />
the effect of climatic fluctuations on the extent and<br />
depth of permafrost, as well as the amount of increase of<br />
groundwater through talik formation on local divides during<br />
phases of climatic warming. The appearance of these<br />
taliks considerably effect the groundwater and hydrological<br />
regime of local rivers, especially in the winter period.<br />
73
Kasymskaya, M. et ai.: Evolution of Alpine permafrost under the impact..<br />
- 1 I<br />
300 ' I<br />
0 5 10 15 <strong>20</strong> <strong>25</strong> 30 35 40 45 50 55 60 65 70<br />
1- surface of relief Number of blocks<br />
-2- permafrost lower boundary, 83 Kyr B.P.<br />
-3- permafrost lower boundary, 6 Kyr B.P.<br />
-4- permafrost lower boundary, recent time<br />
1<br />
""<br />
0 5 10 15 <strong>20</strong> <strong>25</strong> 30 35 40 45 50 55 60 65 70<br />
_-"x 1- surface of relief Number of blocks<br />
-5- permafrost lower boundary, 105 Kyr B.P.<br />
-6- permafrost lower boundary, 22 Kyr E.P.<br />
Fig.3. Surface of relief (upper boundary of alpine permafrost model)<br />
and position of permafrost lower boundary in different time periods.<br />
REFERENCES<br />
Sergueev,D., Tipenko, G., Romanovskii, N., Romanovsky, V., and<br />
Kasymskaya, M. <strong>20</strong>02. Impact of Mountain Topography and Altitudinal<br />
Zonation on Alpine Permafrost Evolution and Ground Water<br />
Hydrology, AGU <strong>20</strong>02 Fall Meeting Abstracts, San Francisco,<br />
California, USA.<br />
Sergueev,D., Tipenko, G., Romanovsky, V., and Romanovskii, N.<br />
Mountain permafrost thickness evolution under influence of longterm<br />
climate fluctuation (results of numerical simulation). <strong>Proceedings</strong><br />
of the 8th International Conference on Permafrost (In<br />
press).<br />
74
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
An estimate of organic carbon inputo the Arctic seas from permafrost and<br />
soils of the East Siberian coasts<br />
A.L. Kholodov, D. E. Fyodorov-Davidov, E. M. Rivkina, S. V. Gubin, V.A.Sorokovikov, L.A. Fominikh, and D.A. Gilichinsky<br />
institute of Physicochemical and Biological Problems in Soil Science, Russian Academy of Science, Pushchino, Russia<br />
Permafrost is a significant reservoir and potential source<br />
of ancient organic matter (plant remains, humified organics,<br />
etc.) and greenhouse gases (C02 and CH4). Due to<br />
the degradation of permafrost as a result of coastal erosion,<br />
thermokarst formation and thawing of submarine<br />
permafrost, this organic carbon is easily released and reabsorbed<br />
again into the present biogeochemical cycle.<br />
The task of this study is to estimate possible organic input<br />
into the Laptev and East Siberia seas and atmosphere<br />
due to these processes as one objective of the IPA Arctic<br />
Coastal Dynamics program.<br />
Investigations were carried out at three key sites: Bykovsky<br />
peninsula (129030'E, 71040'N.), Cape Svyatoi Nos<br />
(14O00'E 7<strong>20</strong>55'N), both in the Laptev Sea region, and<br />
Cape Chukochyi (159-55'E 70005'N), East Siberian Sea<br />
coast. All sites are characterized by the widespread occurrence<br />
of fine-grained Quaternary sediments with large<br />
quantities of organic matter and biogases. There are two<br />
main types of landscapes associated with the deposits.<br />
One is the late Pleistocene (I-alQIII) accumulative plain<br />
(Edoma formation or ice-complex) composed of syncryogenic<br />
ice-rich sediments. Mean annual ground temperatures<br />
within this landscape are -11 to -14OC. Shore cliffs<br />
are <strong>20</strong> to 40 m in height. The other type of landscape is<br />
characterized by alas depressions resulting from the<br />
thawing of ice-complex under thermokarst lakes. The alas<br />
complex includes layers of Holocene lake-boggy deposits<br />
(I-bQlv); layers of tabular deposits (thawed and redeposited<br />
ice complex; tbQlll-lv), and residual ice complex<br />
(Fig. I). Mean annual ground temperature is -9OC and the<br />
height of Laptev shore cliffs is less than 10 m but can<br />
reach <strong>20</strong> m along the East Siberian coast.<br />
One of the main processes in this region is coastal<br />
erosion. The length of eroding shores of the Laptev Sea is<br />
1000-1300 km, and for the East-Siberian Sea up to <strong>20</strong>00<br />
km. The rate of coastal retreat is 2-6 mlyear (Are 1998).<br />
According to data obtained for the Cape Chukochyi at the<br />
East Siberian Sea, the coast has retreated at an average<br />
rate of 4.5 mlyear during the past 18 years. Another im-<br />
portant process which takes place in the region is erosion<br />
of the sea floor. This process is most active on the shallow<br />
shelves near the coast and on the seabed of former is-<br />
lands that were destroyed by erosion over the past centuries.<br />
Total content of organic carbon was determined on dry<br />
soil samples using the wet oxidation method. Volumetric<br />
total Corg content for massive frozen soil was calculated on<br />
the basis of an assumed soil density (1700 kglm3). Calculations<br />
were carried out using a volumetric ice content<br />
of 75% for the ice complex and of 50% for the alas complex.<br />
w C org, YO W C org, %<br />
A 0 2 4 048lZ El 0 2 4 O B *<br />
Figure 1. Texture, water and carbon contents of ice complex deposits<br />
(A) and alas complex (B) on Bykovsky Peninsula.<br />
Supply of Corg for the different relief forms was determined<br />
for the active layer at the CALM sites R13, and<br />
R13A (Brown et al. <strong>20</strong>00). Thickness of the active layer<br />
was taken as 0.5 m for hummocks, 0.4 for spot medallions,<br />
0.3 m on the slope of hummocks, and 0.1 m in<br />
cracks. Soil density was assumed as 150-<strong>20</strong>0 kglm3 for<br />
organic horizons, 800 kglm3 for mineral horizons in the<br />
upper part of the profile, and I700 kglm3 for the lower part<br />
(Fig. 2). Results of the calculation of total Corg in frozen<br />
deposits and modern soils are given in Table 1.<br />
It was determined that 1 m3 of ice complex contains 5-<br />
12 kg of Corg Due to erosion of 1 m2 of coastal territory, an<br />
amount of <strong>25</strong> to 4<strong>25</strong> kg of Corg can be delivered into Arctic<br />
seas.<br />
Deposits of the alas complex are characterized by a<br />
similar total Corg (8-12 kglm3). The main part of organic<br />
75<br />
a<br />
16 '$
Khuladav, A. L. et 31.: Estimate of organ carbon input, Arctic seas. East Siberian coasts,,,<br />
supply (75%) in these deposits is concentrated in upper On the widespread marine terraces, river deltas and<br />
parts of the section which are composed of lake-bogy alas depressions of Northeast Russia, where the bluff<br />
sediments. Tabular deposits have less organic matter<br />
relative to the initial ice complex. Average organic content<br />
in a 50m thick section of massive ice complex on the Bykovsky<br />
peninsula is 22 kglm3, but a 10 m section of tabular<br />
deposits contains 3 to IOkglm3. Thus, due to thawing of<br />
the ice complex during development of the thermokarst<br />
lake, decay of up to 50% of the organic matter took place<br />
(Fig. 1).<br />
height is not more than 5 m, the total Corg input to the Arctic<br />
seas from modern soils is comparable with the Corg<br />
from cliffs. Total amounts of Corg, delivered to the Arctic<br />
seas due to coastal erosion of the Laptev and East Siberian<br />
seas can be up to 0.8 million tonslyear or more, including<br />
0.01 million tons of modern vegetation, 0.03 and<br />
more of modern soil, and more than 0.75 buried in permafrost.<br />
These values are comparable with the input from<br />
river discharge (0.85 million tonslyear) (Danyushevsky<br />
1990).<br />
Due to the erosion of marine banks, the organic carbon<br />
could also be released into the atmosphere. The concentration<br />
of CH4 in permafrost varies from 2 to 40 mllkg and<br />
of C02 from I to <strong>20</strong> mllkg; it should be noted that catbon<br />
dioxide is ubiquitous while the occurrence of methane is<br />
determined by the genesis of the deposits and by the type<br />
of cryogenic stratum. For example, the ice complex on the<br />
East Siberian coast is free of methane, while the alas<br />
complex contains <strong>20</strong> to 30 mllkg of CH4. We have calculated<br />
the amount of methane which could be released after<br />
thawing and abrasion of the Cape Chukochyi alas outcrops<br />
(with sediments containing 30 ml CH4lkg and<br />
outcrop sizes being up to I km in length, - up to <strong>20</strong> m in<br />
Figure 2. Typical profile of modern soil (Cape Chukochyi).<br />
height and with coastal retreat amounting to -1 m). The<br />
resulting value is close to 0.4 tonnes.<br />
Table 1. Content of organic carbon in frozen deposits and modern soil<br />
of Laptev and East Siberian coastal zone.<br />
Borehole Deposits Average<br />
ACKNOWLEDGEMENTS<br />
TOC, kglm3<br />
Bykovsky peninsula<br />
These investigations were carried out as a part of Russian<br />
1-01 I-alQIII 11.2<br />
Academy of Sciences “World Ocean” program and sup-<br />
2-01 N-l-bQIII-lv 12.2<br />
ported by the Russian Foundation for Basic <strong>Research</strong><br />
Svyatoy Nos cape<br />
(<strong>20</strong>03). The modern soil depths of thawing and organic<br />
3-01 I-aIQ~~.~~~ 5.5<br />
matter content were studied within the framework of the<br />
4-01 QII-III 3.4<br />
Chukochyi cape<br />
NSF-supported CALM project (Brown et al. <strong>20</strong>00).<br />
2-91 mQI 71 .I<br />
3-91 IQIII-IV 8.2<br />
4-91 mQtl-rll 8.6<br />
5-91 IQIII-IV 11.2<br />
REFERENCES<br />
24.7<br />
Are F. 1998. The contribution of shore thermo-abrasion to the Laptev<br />
Sea sediment balance. In <strong>Proceedings</strong> of the 7th Intern. Confer-<br />
10-91 I-alQIII 5.9<br />
ence on Permafrost. Yellowknife, Canada: <strong>25</strong>-31,<br />
Modern soil (Chukochyi cape)<br />
Brown J., Hinkel K.M. and Nelson, F.E. <strong>20</strong>00. The Circumpolar Active<br />
Edorna Hummock 19.6<br />
Layer Monitoring (CALM) program: <strong>Research</strong> designs and initial<br />
Spot medallion 14.8<br />
results. Polar Geography 24 (3): 165-<strong>25</strong>8.<br />
Hummock slope 24.5<br />
Danyushevsky A. (ed) 1990. Organic substance of bottom sediments<br />
Crack 5.6<br />
in polar zone of World Ocean. Leningrad, “NEDRA (in Russian).<br />
Alas Hummock 21.7<br />
6-91<br />
8-91 1Q11t.n<br />
IQIII-IV<br />
11.0<br />
Spot medallion <strong>25</strong>.3<br />
Hummock slope 23.1<br />
For the modern soils of the Edoma surface of Cape<br />
Chukochyi, the content of Corg is in the range of 5-9 kglm3.<br />
For alas depressions it is 7.5-11.5 kglm3. Similar values of<br />
Corg for sites along the Bering Sea coast (Svyatoi Lavrenty<br />
Cape) were obtained by D. Zamolodchikov (see abstract<br />
in this volume for site discussion). The average supply of<br />
total Corg in modern vegetation is about I kglm2.<br />
76
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available l n ~ o ~ ~ ~ ~<br />
Long-term monitoring of borehole temperatures and permafrost-related<br />
data for climate change research and natural hazard management: examples<br />
from the Mattertal, Swiss Alps<br />
L. King, R. Hot T. Hen and *S. Gruber<br />
lnstitut fur Geographie, Justus Liebig-Universitat Giessen, Germany<br />
"Glaciology and Geomorphodynamics Group, Department of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
INTRODUCTION<br />
Mattertal and neighbouring Saastal are certainly amongst the<br />
best investigated areas in the Alps with regards to permafrost<br />
distribution and natural hazard management. An extremely<br />
steep topography and traffic routes to touristic sites (e.g.,<br />
Zermatt, Saas Fee) have led to intense research efforts by<br />
transport authorities and scientists in order to minimize risks<br />
for the population and tourists and to acquire the data necessary<br />
for natural hazard assessment and protection. Moreover,<br />
long-term temperature monitoring in deep boreholes and<br />
continuous climate measurements provide valuable data for<br />
permafrost distribution modelling and climate change research.<br />
BOREHOLE MONITORING<br />
RESEARCH<br />
FOR CLIMATE CHANGE<br />
Three boreholes have provided us with significant data concerning<br />
subsoil rock temperatures. The locations and the<br />
characteristics of the two boreholes on the Stockhorn plateau<br />
and the borehole at Ritigraben are shown in Table 1. The<br />
temperature curves vary considerably, mainly due to their<br />
respective positions and surface structures. Permafrost thickness<br />
is expected to be approximately 170 meters (Stockhorn<br />
Plateau) and more than 30 m (Ritigraben), respectively. The<br />
temperature curves show evidence of climatic warming during<br />
the last decades, however, the separation of topographic<br />
from transient effects in complex topography is a challenging<br />
task. The two boreholes at the east-west running crest of the<br />
Stockhorn show a very strong effect of exposure to permafrost<br />
characteristics. On the plateau and close to the northem<br />
face of the ridge -2,5" C have been measured at 30 metres<br />
depth, however, only -0,9" C were recorded at the southern<br />
face.<br />
The Stockhorn and Ritigraben boreholes provide data for<br />
long-term monitoring of permafrost temperatures, an important<br />
source of data for climatic change research. In the high<br />
mountain areas of Europe, this idea was initiated by the E U<br />
project PACE (Permafrost And Climate in Europe) and the<br />
network is maintained by its ESF-sponsored continuation,<br />
PACE-21. The data is also integrated into international or na<br />
tional databases as GTN-P or PERMOS (PERmafrost<br />
Monitoring <strong>Switzerland</strong>), forming a basis for global comparative<br />
research.<br />
CLIMATE MONITORING AND PERMAFROST<br />
MODELLING<br />
Long-term data gathered by the Inter-cantonal Measurement<br />
and Information System (IMIS, ENET) for avalanche warning<br />
provides permafrost-related information such as air and<br />
ground temperatures, precipitation, snow depth, global radiation,<br />
wind speed and wind directionintheMattertal.These<br />
stations are ofien located in permafrost areas, as the 0°Cisotherme<br />
in the Mattertal lies slightly above 2600 m a.s.1.<br />
(Tab. 2).<br />
In connection with information taken from DTMs, aerial<br />
photography, satellite imagery and research undertaken at the<br />
Giessen Institute for Geography (Gruber <strong>20</strong>00; Philippi et al.,<br />
in press), this climate information forms an excellent database<br />
forfurtherpermafrostmodelling (Gruberand Hoelzle <strong>20</strong>01)<br />
using a Geographical Information System (GIs). A field<br />
check with BTS measurements was done in March <strong>20</strong>03. It<br />
is intended to implement this data into a neuronal network as<br />
this could prove to be a valuable tool for a better modelling d<br />
permafrost.<br />
MONITORING FOR NATURAL HAZARD<br />
MANAGEMENT<br />
The strong variability of climatic parameters in this relatively<br />
dry andcontinentalclimate may lead to extreme climatic<br />
events that could trigger catastrophic mass wasting<br />
processes due to the extremely steep topography of the<br />
whole Valais region. The increasing frequency of natural hazard<br />
events are a matter of great concern to the local, regional<br />
and national authorities. Great attention is therefore paid to the<br />
assessment and management of natural hazards.<br />
More and more, it is realized that permafrost plays an important<br />
direct or indirect role in the assessment and prevention<br />
77
of these hazards (Hen et al. in press). The responsible<br />
authorities and research organizations therefore support ap<br />
plied research of the mudflow origins in permafrost areas and<br />
the origin of unexpected water releases, the installation d<br />
early warning devices in mudflow canyons, the monitoring d<br />
degrading permafrost and the examination of the stability d<br />
avalanche protection structures in permafrost areas.<br />
Table 1. Borehole Stockhorn Plateau and Ritigraben, location and<br />
borehole characteristics.<br />
Stockhorn Ritigraben<br />
Co-ordinates<br />
Altitude<br />
Slope inclination<br />
Date of Drilling<br />
Borehole depth<br />
Permafrost thickness<br />
Active layer depth<br />
ZAAIMAGT<br />
Active layer<br />
45"59'00" N<br />
07"49'05" E<br />
3410 m a.s.1.<br />
8" S<br />
August <strong>20</strong>00<br />
100 m I30 m<br />
170 m (approx.)<br />
3.1 m<br />
17.7 m/-2,5"C<br />
bedrock<br />
Temperature [ i C]<br />
King, L. et al: Barehole temperatures and permafrast-related data.,.<br />
46" 10' 27" N<br />
07" 50' 56" E<br />
2615 m a d.<br />
14" NW<br />
October <strong>20</strong>01<br />
30 m<br />
>30 m<br />
3.8 m<br />
14 m/-0,37"C<br />
boulders<br />
-10 -8 -6 -4 -2 0 2 4 6 8 10<br />
0<br />
Table 2. Selected IMlS and ENET-Climatic stations used for modelling<br />
purposes in the Matter valley.<br />
Name<br />
<strong>20</strong>01<br />
Altitude MAAT<br />
Gornergrat snow station 2950m -2,l "C<br />
Gornergrat wind station 31 30m -2,7"C<br />
Saas Platthorn 3246m -3,9"C<br />
Saas Schwarzmies 281 Om -1,3"C<br />
Saas Seetal 2480m +0,5"C<br />
St. Niklaus Ob. Stelli Glacier 291 Om -1,6"C<br />
Zermatt Platthorn 3345m -4,5"C<br />
Zermatt Triftchumme 2750m -0.6"C<br />
CONCLUSIONS<br />
Long-term monitoring of permafrost temperatures in deep<br />
boreholes and the analysis and interpretation of climate data<br />
forms a valuable source of information for climate change re<br />
search and permafrost distribution modelling. This information<br />
can also be used for natural hazard detection and protection.<br />
10<br />
REFERENCES<br />
- e<br />
<strong>20</strong><br />
30<br />
40<br />
1 50<br />
60<br />
70<br />
80<br />
, , -<br />
. . . . . . . . .<br />
, ,<br />
Maximum temperature<br />
Gruber, S. <strong>20</strong>00. Slope instability and permafrost - a spatial analysis<br />
in the Matter valley, <strong>Switzerland</strong>. - unpublished diploma<br />
thesis, Institute for Geography, University of Giessen, Germany.<br />
Gruber, S. and Hoelzle, M. <strong>20</strong>01. A statistical modelling of mountain<br />
permafrost distribution: a local calibration and incorporation of<br />
remotely sensed data. Permafrost and Periglacial Processes 12<br />
(1): 69-77.<br />
Hem, T., King, L. and Gubler, H. (in press). Microclimate within<br />
coarse debris of talus slopes in the alpine periglacial belt and<br />
its effects on permafrost. 8th International Conference on Permafrost,<br />
<strong>Proceedings</strong>, <strong>Zurich</strong>, 19-<strong>25</strong> <strong>July</strong> <strong>20</strong>03.<br />
Philippi, S., Herz. T. and King, L. (in press): Near-surface ground<br />
temperature measurements and permafrost distribution at Gornergrat,<br />
Matter valley, Swiss Alps. - Poster Abstract. - In: International<br />
Conference on Permafrost, <strong>Proceedings</strong>, <strong>Zurich</strong>, 19-<br />
<strong>25</strong> <strong>July</strong> <strong>20</strong>03.<br />
90<br />
100<br />
-0,005 0 0,005 0,Ol 0,015 0,02 0,0<strong>25</strong><br />
Temperature gradient [ iCh]<br />
Figure 1. Rock temperatures and thermal gradient of the 100 meter<br />
deep Stockhorn borehole (01-10-<strong>20</strong>01 to 30-09-<strong>20</strong>02).<br />
An early warning device has been installed in the Ritigraben<br />
gully above St. Niklaus. In the event of a debris flow it<br />
would automatically cause the main road in the valley to be<br />
closed. In a further step it is intended to use precipitation data<br />
from the weather station<br />
the Ritigraben borehole as an early<br />
warning parameter, that is located close to the starting point d<br />
the debris flows.<br />
78
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available 1~~~~~~~~~~<br />
Destruction of coasts on the Yugorsky Peninsuland on Kolguev Island,<br />
Russia<br />
A. 1. Kizyakov and * D. D. Perednya<br />
Earfh Cryosphere lnsfitufe<br />
SB RAS, Tyumen, Russia<br />
*Moscow State University, Faculty of Geography, Russia<br />
Cryolithological, environmental and climatic features which<br />
determine coastal processes were investigated at the western<br />
coast of the Island Kolguev (Barents Sea) in summer <strong>20</strong>02<br />
and at the central part of the Yugorsky Peninsula coast (Kara<br />
Sea) in the summers of 1999,<strong>20</strong>01 and <strong>20</strong>02 (Cherkashov et<br />
al. 1999, Kizyakov et al. <strong>20</strong>02, Perednya et al. <strong>20</strong>02).<br />
Coasts at both sites consist of frozen sandy-clayey deposits<br />
enclosing tabular (massive) ground ice. Such tabular ice<br />
bodies control the formation of thermodenudational cirques as<br />
characteristic landforms: such thermocirques develop in<br />
slopes where tabular ground ice is found near the surface.<br />
They are formed by a variety of processes: thermokarst,<br />
coastal and lateral thermoerosion, retrogressive thaw slumps<br />
and earth Rows.<br />
The observed key sites d&r in the activity of hydrodynamic<br />
processes. The western coast of Kolguev Island is<br />
open to wave action and the dynamically active ice-free ps<br />
riod at the Barents Sea is longer than in the Kara Sea. At the<br />
Yugorsky key site, the hydrodynamic processes are weaker<br />
as a result of a very short ice-free period: the Kara Sea is<br />
completely covered with ice from October to May. Drift ice<br />
stays practically all summer. These conditions determine the<br />
character and dynamics of coastal destruction at the study<br />
sites.<br />
Coastal bluffs on Kolguev Island are up to 40-50 m high.<br />
The coasts are of thermoerosion type, directly affected by<br />
waves. Bluffs are steep to overhanging. At the slope foot<br />
wave-erosion niches are formed and often remain partially<br />
filled with snow. The beach is narrow, up to 10 m wide and<br />
the bluff edges are dissected by numerous gullies which reflect<br />
a polygonal pattern (Fig. I),<br />
The three investigated coastal thermocirques are <strong>20</strong>0-400<br />
m long (along the coast), and up to <strong>20</strong>0 m wide (landward).<br />
Analysis of the shoreline is done based on field inspection<br />
and aerial images from 1948 and 1968. Coastal retreat averages<br />
3-4 mlyear (Perednya et al. <strong>20</strong>02).<br />
One of the thermocirques formed after 1968. The retreat<br />
rate of its scarp was therefore no less than 5 mlyear. Older<br />
thermocirques, however, are much less active and most<br />
likely retreat at a rate of less than1 mlyear. At the Yugorsky<br />
key site, coastal bluffs have relative heights from 12 up to 29<br />
m. Coasts are of the thermoerosion-thermodenudation type<br />
narrow beach (I), s~ee~ to ovefh<br />
w~ve-erod~d ni~h~s (3) at the slope f~~~ are<br />
urnerou~ ~ullies at the b i ed ~ ~<br />
~a~t~rns.<br />
and wave-erosion niches are not formed (Fig. 2). A complex<br />
of thermodenudation slope processes destroys coastal bluffs.<br />
Sediments are washed away by tides and most actively by<br />
storm surges after having been deposited on a beach. These<br />
sediments are involved in a shore-parallel drift directed westtoeast.<br />
In <strong>20</strong>01, a monitoring network was established at a<br />
thermocirque scarp edge. The dynamics of long-term coastal<br />
retreat and thermocirque growth was recorded based on a<br />
comparison of field and remote-sensing data. Interpretation d<br />
aerial images from 1947 was combined with an analysis d<br />
topographic maps compiled using images from 1967.<br />
The topographic survey of thermocirques and coastal-bluff<br />
edges was performed by theVNIIOoceangeology Institute,<br />
St-Petersburg in <strong>20</strong>01 (Kizyakov <strong>20</strong>02, Kizyakov et al.<br />
<strong>20</strong>02). The scarp most actively retreating from 1947 to <strong>20</strong>01<br />
was analysed and the retreat rate determined as 0.6 to 1<br />
mlyear, while the coastal-bluff retreat amounted to 1.3 - 1.6<br />
mlyear. The topographic survey enabled the compilation of a<br />
digital terrain model in SURFER software, which provided a<br />
reconstructed topography of the area prior to thermocirque<br />
formation and formed the basis forcal culating the sediment inputfromthe<br />
thenocirque in 54 years. According to these<br />
calculations (Tab. I), the thermocirques on the Yugorsky<br />
79
coast increase the sediment yield, because the scarp edges<br />
are curved and thus longer than the shoreline for the smoothfaced<br />
bluffs.<br />
Kizyakov, A. I. et al.: Destruction of coasts an the Yugorsky Peninsula<br />
ACKNOWLEDEMENT<br />
The study was performed within the framework of INTAS the<br />
projects, grants 01-2329 and 01-2211.<br />
REFERENCES<br />
The corresponding sediment flux from a vast catchment area<br />
is concentrated through a narrow exit. On average, sediment<br />
input from thermocirques appeared to be approximately 3<br />
times as high as from the smooth-faced coastal bluffs. The<br />
total input of thermocirques for a large portion of the coast is<br />
rather low because the latter occupy only about 6% of the<br />
Yugorsky shoreline.<br />
Cherkashov, G.A., Goncharov, G.N., Kizyakov, A.I., Krinitsky, P.I.,<br />
Leibman, M.O., Persov, A.V., Petrova, V.I., Solovyev, VA., Vanshtein,<br />
B.G. and Vasiliev, AA. 1999. Arctic coastal dynamics in<br />
the areas with massive ground ice occurrence. Report of an<br />
International Arctic Coastal Dynamics Workshop, Woods Hole,<br />
2-4 November 1999. Geological Survey of Canada Open File<br />
3929.<br />
Kizyakov, A.I., Perdnya, D.D., Firsov, Yu.G., Leibman, M.O. and<br />
Cherkashov G.A. <strong>20</strong>02. Character of the costal destruction and<br />
dynamics of the Yugorsky Peninsula coast. Arctic Coastal Dynamics<br />
3rd Workshop <strong>Proceedings</strong>, Oslo, 2-5 December <strong>20</strong>02,<br />
(in press).<br />
Kizyakov A.I. <strong>20</strong>02. The forms of Kara Sea coasts destruction in<br />
conditions of widespread tabular ground ice. Extreme phenomena<br />
in cryosphere: basic and applied aspects. Abstracts of<br />
the International Conference, Pushchino, 12-1 5 May <strong>20</strong>02.<br />
Pushchino: ONTl PNC RAS.<br />
Perdnya, D.D., Leibman, M.O., Kizyakov, A.I., Vanshtein, B.G. and<br />
Cherkashov, G.A. <strong>20</strong>02. Coastal dynamics at the western part of<br />
Kolguev Island, Barents Sea. Arctic Coastal Dynamics 3rd<br />
Workshop <strong>Proceedings</strong>, Oslo, 2-5 December <strong>20</strong>02, (in press).<br />
Table 1. The rate of retreat due to coastal thermoerosion and<br />
modenudation at the Yugorsky Peninsula and sediment yield<br />
km of the shoreline<br />
therper<br />
Destructive Retreat rate Sediment yield<br />
processes m per year m3 per year<br />
Coastal thermoerosion 1.3-1.6 30100<br />
and falls at the coastal<br />
bluffs<br />
Complex of 0.6-1 -04 89000<br />
thermodenudation<br />
processes in thermocirques<br />
Thus, the western coast of Kolguev Island is of the thermoerosion<br />
type, and collapses as a result of direct mechanical<br />
and thermal wave influence. The coast of Yugorsky<br />
Peninsula in its central part, however, is of the thermoem<br />
sion-thermodenudation type and collapses mainly due b<br />
slope thermodenudation processes. At both study sites, thermocirques<br />
develop within smooth-faced slopes by thawing d<br />
tabular ground ice. The coastal retreat on Yugorsky Peninsula<br />
for the smooth-faced bluffs is slower by a factor of two than<br />
onthewesterncoastofKolguevIsland.Activetherrnocirques<br />
of Kolguev Island are developing up to 5 times faster<br />
than on Yugorsky Peninsula. The development of thermocirques<br />
in the coastal zone increases the sediment yield per<br />
metre of a shoreline.<br />
This points toward more active destruction of the Kolguev<br />
site in the western aspect of the study coast of Kolguev Island<br />
and specific hydrodynamic features (longer ice-free pe<br />
riod, more wave action) of the Barents Sea.<br />
80
8th International Conference on Permafrost Extended Abstracts an Current <strong>Research</strong> and Newly Available ~~f~~~~~~~~<br />
New insights of mountain permafrost occurrence and characteristics in<br />
recently deglaciated glacier forefields through the application of electrical<br />
resistivity tomography<br />
C. Kneisel<br />
University of Wuerzburg, Department of Physical Geography, Wuerzburg, Germany<br />
INTRODUCTION<br />
Different geophysical methods have been applied successfully<br />
in mountain regions to detect mountain permafrost. Since<br />
geoelectrical methods are most suitable for investigating the<br />
subsurface with distinct contrasts in conductivity and resistivity,<br />
DC resistivity soundings constitute one of the traditional<br />
geophysical methods which are standardly applied in pemafrost<br />
research to confirm mountain permafrost over many<br />
years. During recent years advances have been achieved in<br />
usingthetraditionalmethodsbut with more powerful instruments<br />
and modem data processing algorithms which enable<br />
two-dimensional sutveys and data processing (Vonder Muhll<br />
et al. <strong>20</strong>01).<br />
The study sites are situated in the Upper Engadine, eastem<br />
Swiss Alps, in the vicinity of St. Moritz (46"30'N,<br />
9'50'E). Numerous rock glaciers are obvious geomorphological<br />
indicators of the discontinous distribution of alpine<br />
permafrost. Various aspects of mountain permafrost have<br />
been investigated in the past decades with the main focus on<br />
rock glacier permafrost. Discontinuous permafrost is encountered<br />
above 2400 m a.s.1. The investigation of permafrost in<br />
glacier forefields at high altitudes in the Upper Engadine was<br />
performed during recent years using onedimensional gm<br />
electrical soundings as the standard geophysical technique b<br />
detect mountain permafrost. The results of numerous measurements<br />
confirmed the local presence of perennially frozen<br />
ground and ground ice in the investigated glacier forefields<br />
(Kneisel 1998, 1999).<br />
On heterogeneous ground conditions the interpretation d<br />
one-dimensional soundings can be difficult, as lateral variations<br />
along the survey line can influence the results significantly.<br />
The sounding curve produces an average resistivity<br />
model of the survey area, but individual anomalies such as<br />
small permafrost lenses will not show explicitly the in results.<br />
Hence, the interpretation is limited with respect to the characteristics<br />
and the local distribution pattern. Recently, twodimensional<br />
electrical resistivity tomography which overcomes<br />
this problem using multielectrode systems and twodimensional<br />
data inversion was applied in the same glacier<br />
forefields and at the same location confirming the assumptions<br />
made by the one-dimensional vertical soundings.<br />
The motivation of this contribution is to illustrate the opportunities<br />
and potentials of two-dimensional DC resistivity b<br />
mography for periglacial geomorphology providing a more<br />
detailed characterization of the subsurface as compared to the<br />
results obtained by traditional one-dimensional vertical<br />
soundings.<br />
METHODS<br />
For the one-dimensional geo-electrical soundings a GGA30<br />
Bodenseewerk instrument was used. Three-layer models<br />
were calculated which are assumed to represent the most<br />
probable case with respect to the local geomorphological<br />
situation. For the interpretation of one-dimensional data the assumption<br />
is made that the subsurface consists of horizontal<br />
layers and that the resistivity changes only with depth but not<br />
horizontally. The typical one-dimensional geeelectrical<br />
sounding curve obtained on alpine permafrost shows a three<br />
layer model with an increase of resistivity at shallow depth<br />
representing the active layer and the permafrost underneath,<br />
followed by a sharp decrease of resistivities greater current<br />
electrode distances representing the unfrozen ground beneath.<br />
For the two-dimensional electrical tomographies a SYSCAL<br />
Junior Switch system was applied. The surveys were conducted<br />
using the Wenner configuration.<br />
Resistivity values of frozen ground can vary over a wide<br />
range depending on the ice content, the temperature and the<br />
content of impurities. The dependance of resistivity on tem-<br />
perature is closely related to the amount of unfrozen water.<br />
Perennially frozen silt, sand, gravel or frozen debris with<br />
varying ice content show a wide range of resistivity values<br />
between 5 kWm to several hundred kWm (Haeberli and<br />
Vonder Miihlli996).<br />
81<br />
RESULTS<br />
The resistivity sounding profile shown in Figure 1 was measured<br />
in a high altitude glacier forefield using the symmetrical<br />
Schlumberger configuration. The graph shows only a slight
Kneisel. C,: Mountain permafrost, glacier forefields, electrical resistivity tomography,.,<br />
increase of the resistivity values. A three-layer model inversion<br />
was performed suggesting a first layer of 3 to 7 m representing<br />
the active layer and a second layer with a thickness<br />
of about IOto <strong>25</strong> m. The resistivity values obtained for this<br />
middle layer (between <strong>20</strong> km and 35 km for equivalent models),<br />
lie in the range of perennial frozen ground and suggest a<br />
shallow permafrost occurrence at this site. The third layer<br />
with better conductivityrepresents the unfrozen ground underneath.<br />
Figure 2. Resistivity tomogram of a Wenner<br />
permafrost<br />
survey with mountain<br />
B<br />
E<br />
Figure 1. One-dimensional DC resistivity sounding profile<br />
I<br />
The described examples from two surveys in the periglacia1<br />
environment illustrate that electrical resistivity tomograph<br />
is a powerful tool to determine the structure of the subsurface<br />
at different resolutions, depending on the electrode spacing<br />
and array geometry employed. For a reliable interpretation d<br />
the survey results, knowledge of the geomorphological and<br />
geological setting is essential. Further details of applicability d<br />
different array geometries, advantages and disadvantages d<br />
the electrical resistivity tomography and its application b<br />
various geomorphological studies are given in Kneisel<br />
(<strong>20</strong>03).<br />
REFERENCES<br />
This shape of geoelectrical sounding graphs is considered<br />
to be typical of permafrost in glacier forefields. Similar curves<br />
have also been measured in other investigated forefields.<br />
With the special conditions in glacier forefields, permafrost OG<br />
currences can be assumed to be shallow, "warm" and thus,<br />
rich in unfrozen water, what can explain the comparatively<br />
low apparent resistivity values (Kneisel 1999). Figure 2 illustrates<br />
the results of the two-dimensional resistivity survey<br />
performed at the same location where the one-dimensional<br />
geoelectrical soundings were performed in the previous study<br />
which already indicated a permafrost occurrence. Thedark<br />
grey colors represent perennially frozen ground. Through the<br />
onedimensional sounding similar thickness of the active<br />
layer and depth of the permafrost were obtained. However,<br />
since one-dimensional geoelectrical soundings produce only<br />
an average resistivity model of the survey area, the lenticular<br />
high-resistive areas as shown in the 2-D image could not be<br />
deduced.<br />
Additionally, topography was incorporated in the inversion,<br />
resulting in a more realistic interpretation of the obtained<br />
resistivity image of the subsurface. This is especially evident<br />
in the depression between station 45 and 55 where the active<br />
layer is markedly thinner due to the late-lying snow. Here,<br />
the potential of electrical resistivity tomography for geomorphological<br />
investigations compared to onedimensional<br />
soundings is shown clearly, as a more sophisticated interpretation<br />
of the subsurface can be derived. Furthermore, the<br />
opportunity for a real mapping over larger areas is a major<br />
advantage for geomorphological studies.<br />
Haeberli, W. and Vonder Muhll, D. 1996. On the characteristics and<br />
possible origins of ice in rock glacier permafrost. 2. f. Geomorph.<br />
N.F. Suppl. 104: 43-57.<br />
Kneisel, C. 1998. Occurrence of surface ice and ground icelpermafrost<br />
in recently deglaciated glacier forefields, St. Moritz area,<br />
Eastern Swiss Alps. 7th International Conference on Permafrost,<br />
Yellowknife, Canada. <strong>Proceedings</strong>: 575-581.<br />
Kneisel, C. 1999. Permafrost in Gletschervorfeldern - Eine vergleichende<br />
Untersuchung in den Ostschweizer Alpen und Nordschweden.<br />
Trierer Geographische Studien 22.<br />
Kneisel, C. in press. Electrical resistivity tomography as a tool for<br />
geomorphological investigations - some case studies. In: L.<br />
Schrott, A. Hoerdt and R. Dikau (eds.). Geophysical methods in<br />
geomorphology. Z. f. Geomorph., Suppl.<br />
Vonder Muhll, D., Hauck, C., Gubler, H., McDonald, R. and Russill,<br />
N. <strong>20</strong>01. New geophysical methods of investigating the nature<br />
and distribution of mountain permafrost with special reference<br />
to radiometry techniques. Permafrost and Periglacial Processes<br />
12: 27-38.<br />
82
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Evidence of paleotemperature signals in mountain permafrost areas<br />
T. Kohl and *S. Eruber<br />
Geophysical Institute, ETH <strong>Zurich</strong>, <strong>Switzerland</strong><br />
*Glaciology and Geomorphodynamics Group, Department of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
INTRODUCTION<br />
In Alpine environments ground temperature changes may<br />
have a significant effect on permafrost. As a consequence,<br />
future warming is likely to lead to increased instability<br />
in mountain permafrost terrain. However, field studies<br />
of thaw-related slope instability are hampered by the<br />
natural variability of geomorphic processes, both in space<br />
and time. To quantify the impact of recent climate variation,<br />
borehole temperature measurements were taken using<br />
30 thermistors at the 3400 m high Stockhorn in the<br />
central part of the Swiss Alps. The measurements show<br />
non-linear thermal profiles, with pronounced near-surface<br />
temperature deviations from the deeper thermal gradient.<br />
Although climatic-induced temperature effects seem to be<br />
evident, other topography-induced thermal signals cannot<br />
be excluded. The present analysis aims at the investigation<br />
of coupled transient and topography effects which<br />
should also elucidate the significance of temperature<br />
measurements in alpine terrains.<br />
temperature distribution and accurate topography representation.<br />
The analysis of these near-surface energy<br />
fluxes is therefore represented by a two-dimensional homogeneous<br />
model assuming constant basal heat flow at<br />
3.5 km depth below the Stockhorn but with transient surface<br />
temperature variation. The model included <strong>20</strong>,000<br />
nodes. The mesh was strongly coarsened from surface<br />
( 4 m distance between nodes) to greater depth (100 m<br />
distance). Topography was taken at IOm distance from a<br />
DTM along a N-S running profile of the Stockhorn.<br />
RESULTS<br />
METHOD<br />
The investigation was conducted using the FE code<br />
FRACTure (Kohl and Hopkirk 1995). One of the features<br />
of this code is the integration of complex topography and<br />
surface condition into a robust and reliable numerical<br />
code. The FE-forward scheme is now linked with the<br />
UCODE computer program which can be broadly used for<br />
various applications. In combination, UCODE and FRAC-<br />
Ture perform inverse modelling, posed as a parameterestimation<br />
problem, by calculating parameter values that<br />
minimize a weighted least-squares objective function using<br />
non-linear regression, With this tool complex multidimensional<br />
inversion studies can be theoretically accomplished.<br />
However, numerical instability and nonuniqueness<br />
represent well-known problems in inverse modelling,<br />
and obtaining useful results can become more difficult for<br />
highly heterogeneous conditions. Therefore, the problem<br />
was focused on properties that seem to have the major<br />
impact on the near-surface temperature field: surface<br />
Figure 1. Temperature profiles along selected sites: the measured<br />
temperature data are given by black dots, the T-profile at x=<strong>20</strong>75m<br />
represents the calculated temperature from steady-state assumptions.<br />
83
only a small displacement of the borehole location would<br />
result in completely different T(z) profiles. Under steadystate<br />
conditions, the results are nearly independent of<br />
thermal conductivity along the borehole depth. To account<br />
for transient effects, a climatic time series from a nearby<br />
study (Bohm et al. <strong>20</strong>01) was initially superimposed on the<br />
surface temperature. When performing 2nd order regression<br />
on these strongly varying data, basically a I .3"C<br />
warming between 1900 and <strong>20</strong>00 can be identified, between<br />
1800 and 1900 surface temperature did not change<br />
drastically. However, our calculations have shown better<br />
fits with a more pronounced warming in the last 50 years.<br />
Figure 2 illustrates the additional curvature in the T(z) profile<br />
due to climatic warming.<br />
locations to optimize different configurations and temperature<br />
logging devices at specific sites.<br />
REFERENCES<br />
Bohm R., Auer I., Brunetti M., Maugeri M., Nanni T. and Schoner W.<br />
<strong>20</strong>01. Regional temperature variability in the European Alps 1760<br />
- 1998 from homogenised instrumental time series. International<br />
Journal of Climate 21: 1779-1801.<br />
Kohl T. and Hopkirk R.J. 1995, "FRACTure" a simulation code for<br />
forced fluid flow and transport in fractured porous rock.<br />
Geothermics 24(3): 345-359.<br />
CONCLUSION<br />
Our analysis has demonstrated that the ground surface<br />
temperature distribution is highly sensitive to the shallow<br />
(-100 m deep) subsurface temperature field. In alpine terrains,<br />
energy fluxes can even be totally dominated by topography<br />
without any influence of basal heat flow. Under<br />
these circumstances, temperature measurements need to<br />
be carefully designed and located at those points which<br />
are less sensitive and which provide the most useful data.<br />
Our procedure can also be applied successfully to further<br />
84
Hydrological modelling of (sub)arctic river systems: responses to climate<br />
change<br />
E. Koster, R. Dankers and S. van der linden<br />
Centre for Geo-ecological <strong>Research</strong>, Department of Physical Geography, Utrecht University, Utrecht, lhe Netherlands<br />
PERMAFROST HYDROLOGY<br />
Hydrologic modelling of river regimes is restricted by our<br />
poor knowledge of permafrost-related processes and by a<br />
paucity of data (Woo <strong>20</strong>00). Frozen ground is relatively<br />
impermeable, especially when ground ice content is high.<br />
Warm spring temperatures melt the bulk of the snowpack<br />
within a few days or weeks, while the soils are still frozen.<br />
The closer to the summer solstice, the greater the incoming<br />
solar radiation so that the spring thaw is often intense<br />
(Woo <strong>20</strong>00), leading to overland flow as the rate of melt<br />
exceeds infiltration capacity. When the active layer thickness<br />
increases towards the end of the thawing season, infiltration<br />
rates increase and surface runoff is correspondingly<br />
reduced. Not surprisingly, the annual catchment<br />
water balance in permafrost areas often shows a high<br />
runofflprecipitation ratio. Typical values range from 0.7-0.8<br />
in high-Arctic polar deserts to 0.4-0.6 in sub-Arctic tundra<br />
regions and wetlands because of higher evaporation (Woo<br />
<strong>20</strong>00). However, wide variations in ratios do occur between<br />
different catchments. During summer with greater<br />
thaw depth, depleted groundwater stores, and increased<br />
evapotranspiration, the hydrograph will respond less<br />
markedly to rainfall. Nevertheless, rainstorms may still occasionally<br />
produce strongly peaked hydrographs. In winter,<br />
river discharge is usually low or sometimes even absent.<br />
Another characteristic of the periglacial fluvial<br />
domain is the sometimes very high percent of the annual<br />
runoff that occurs within the two to three weeks of the<br />
snowmelt flood (break-up), with a corresponding concentration<br />
of sediment transfer. Finally, the presence of permafrost<br />
accentuates runoff peaks while reducing groundwater<br />
supply to river runoff and thereby base flow.<br />
in northern Russia were conducted (Dankers <strong>20</strong>02, Van<br />
der Linden <strong>20</strong>02). The hydrographs of both river systems<br />
fall within the category of the subarctic nival river regime.<br />
The Tana River catchment is 16,000 km* in size; mean<br />
annual discharge is only 166 m3/sec; interannual variation<br />
is high; snowmelt runoff, with a maximum discharge as<br />
high as 3,000 m3/sec, may contribute as much as 65 YO to<br />
the total annual runoff; mean annual T varies from -0.5 to -<br />
3°C and mean annual P from ca 350-450 mm; discontinuous<br />
permafrost occurs. The Usa catchment is 93,000 km*<br />
in size; peak discharges vary from ca 4,000 to more than<br />
7,000 m3lsec; mean annual T varies from -3 to -7°C and<br />
mean annual P from 400 -800 mm; continuous, discontinuous<br />
and sporadic permafrost zones occur from north to<br />
south.<br />
The sensitivity of the discharge of the Tana and Usa<br />
rivers to changes in climate was evaluated with hydrological<br />
models, called TANAFLOW and USAFLOW, respectively,<br />
in both cases based on a spatially-distributed water<br />
balance model (RHINEFLOW) developed for the Rhine<br />
River by Kwadijk (1993). Both models are embedded in a<br />
GIS and calculate the water balance for grid cells of I by I<br />
km; runoff is subsequently routed over a digital elevation<br />
model (DEM) of the area and accumulates at the catch-<br />
ment outlet. Originally designed for monthly time steps,<br />
the models have now been applied on a 10-day basis.<br />
Moreover, the USAFLOW model (Van der Linden <strong>20</strong>02)<br />
has been adapted for cold-climate conditions by introducing<br />
variable (in space and time) separation coefficients -<br />
dividing rapid (surface) runoff from water added to<br />
groundwater storage - for continuous (1.0 100% rapid<br />
runoff), discontinuous (0.9), sporadic permafrost (0.8) and<br />
for seasonal frost (0.6).<br />
WATER BALANCE MODELLING<br />
In the framework of two EC-funded research projects,<br />
“Barents Sea Impact Study” (BASIS) and “Tundra Degradation<br />
in the Russian Arctic” (TUNDRA), hydrological<br />
modelling studies of the Tana river in northern Fennoscandia<br />
and the Usa River as part of the Pechora basin<br />
IMPACT OF CLIMATE CHANGE<br />
In general, it is expected that should warming occur, permafrost<br />
will degrade, and storage capacity of the soil will<br />
increase. Consequently, there will be fewer high flow<br />
events of large magnitude, and the probability of extremely<br />
low flows will also diminish. RIP ratios will de-<br />
85
crease as groundwater storage increases, but surface flow<br />
remains important during spring (Woo <strong>20</strong>00, Van der Linden<br />
<strong>20</strong>02). By introducing temperature and precipitation<br />
changes concurrent with state-of-the-art climate scenarios,<br />
the sensitivity of river discharge to these climate variables<br />
was established (Harding et al. <strong>20</strong>02). Fig. IA illustrates<br />
the change in the Tana hydrograph between a<br />
control run (simulated for the period 1980-1993) and a<br />
scenario run (simulated for the period <strong>20</strong>70-2100), derived<br />
from the RCM HIRHAM4 model. The combined effects of<br />
increased temperatures, precipitation and evaporation result<br />
in a distinctly earlier timing (by about <strong>20</strong> days) of the<br />
runoff peak, which is also slightly higher. The latter is due<br />
to the somewhat larger snow mass, in spite of the shorter<br />
snow season, and reduced sublimation. Although smaller<br />
in amounts, the relative increase in runoff is even larger in<br />
the rest of the year. As baseflow is higher, runoff in winter<br />
has increased in the scenario run as well. Fig. IB illustrates<br />
the change in the Usa hydrograph for the present<br />
situation and according to a GCM HADCM2 scenario projection<br />
to the period around <strong>20</strong>80. It shows a decrease in<br />
peak discharge. In this region the increased evaporation in<br />
summer due to a rise in temperature is not compensated<br />
by larger precipitation amounts, which causes discharge<br />
to decrease. To further improve hydrological models, the<br />
feedback effects of vegetation responses to climate<br />
change should be taken into account. In conclusion, these<br />
studies reveal that the indirect effects of climate change<br />
(e.g., vegetation and permafrost changes) on discharge in<br />
(sub)arctic areas are relatively small compared to direct<br />
effects.<br />
Koster, E. el al.: Hydrological modelling, climate change ...<br />
REFERENCES<br />
Dankers, R. <strong>20</strong>02. Sub-arctic hydrology and climate change. A case<br />
study of the Tana River Basin in Northern Fennoscandia. Netherlands<br />
Geographical Studies 304,237 pp.<br />
Harding, R., Kuhry, P., Christensen, T.R., Sykes, M.T., Dankers, R.<br />
and Van der Linden, S. <strong>20</strong>02. Climate feedbacks at the tundrataiga<br />
interface. Ambio Special Report 12: 47-55<br />
Kwadijk, J.C.J. 1993. The impact of climate change on the discharge<br />
of the River Rhine. Netherlands Geographical Studies 168, <strong>20</strong>1<br />
PP.<br />
Van der Linden, S. <strong>20</strong>02. Ice rivers heating up. Modelling hydrological<br />
impacts of climate change in the (sub)arctic. Netherlands Geographical<br />
Studies 308, 174 pp.<br />
Woo, M-K. <strong>20</strong>00. permafrost hydrology. in: Nuttall, m. and Callaghan,<br />
T.V. (eds.): The Arctic, Environment, People, Policy. Harwood,<br />
Amsterdam: 57-96<br />
86
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Using BTS for mapping permafrost distribution in Apussuit, SW Greenland<br />
L. Krisfensen<br />
Institute of Geography, University of Copenhagen, Denmark<br />
INTRODUCTION<br />
Permafrost in Greenland has been mapped on a national<br />
scale by Weidick (1968) and Christiansen and Humlum<br />
(<strong>20</strong>00) but no previous1 attempts have been made to map<br />
mountain permafrost on a local or regional scale. Geomorphological<br />
indicators of permafrost such as pingos and<br />
rock glaciers, as well as borehole temperatures from permafrost,<br />
are reported from various locations in Greenland.<br />
On the west coast of Greenland the southernmost location<br />
of reported geomorphological indicators of permafrost is in<br />
the Sisimiut area. Here, borehole temperatures show thin<br />
permafrost at sea level.<br />
The BTS Method (Haeberli 1973) has been succesfully<br />
used for mapping mountain permafrost in the Alps and<br />
other mountain regions but, so far, has not been used in<br />
Greenland.<br />
SITE DESCRIPTION<br />
The Apussuit study area covers 30 * 40 km and is located<br />
on mainland Greenland at 65”32’N, 52’22’E, <strong>25</strong> km NE of<br />
Maniitsoq, approximately 30 km from the outer west coast<br />
and 100 km west of the Inland Ice sheet. The area is 10-<br />
cated in a zone of discontinuous permafrost and on the<br />
border between sporadic and discontinous permafrost on<br />
Weidick’s (1968) and Christiansen and Humlum’s (<strong>20</strong>00)<br />
permafrost maps of Greenland. However, no previous<br />
studies on permafrost have been carried out in the area.<br />
The Apussuit ice cap sits on a narrow plateau at 800 -<br />
1133 m a.s.1. and measures around 110 km2. Some of the<br />
highest mountains of western Greenland are located north<br />
of Apussuit, where large ice caps can also be found. The<br />
mountains were probably nunataks during the Quartenary<br />
glaciations (Weidick 1968). The landscape south of Apussuit<br />
has a gentler, glacially-eroded topography. Sediments<br />
are sparse due to low weathering rates of the archaic<br />
gneisses, so apart from block fields and talus slopes,<br />
geomorphological indicators of permafrost are rare. The<br />
fieldwork was carried out near a small skisport center at a<br />
nunatak in 931 m a.s.1. The mean annual air temperature<br />
at sea level is around -1°C.<br />
METHOD<br />
A Digital Terrain Model (DEM) with a cell size of 40 * 40 rn<br />
was interpolated from 150,000 point values originating<br />
from photogrammetry of aerial photographs. The potential<br />
direct solar radiation (PR) was calculated on an hourly basis<br />
throughout the DEM and averaged over a year. The<br />
calculations include damping of the radiation through the<br />
atmosphere and shadow from the surrounding horizon.<br />
160 BTS measurements at 81 points were recorded<br />
between 9<strong>25</strong> m a.s.1. and sea level, together with GPS location,<br />
snow depth, inclination and aspect. The measurements<br />
were made at snow depths between 85 cm and <strong>20</strong>0<br />
cm. The mean distance between the recorded GPS point<br />
and the centre of the appropriate cell in the DEM was 15.1<br />
m. The differences between measured and modelled inclination<br />
and aspect were 8.4’ and 39.8” respectively. These<br />
differences are most likely caused by differences in scale<br />
as the parameters in the field were measured for an area<br />
covering around IO* 10 m, whereas the cell size in the<br />
model was 40 * 40 m.<br />
RESULTS AND DISCUSSION<br />
I30 measurements were in the “permafrost probable” BTS<br />
class (BTS-3”C), 17 in the “permafrost possible” BTS<br />
class (-3°C BTS -2°C) and 13 in the “permafrost improbable”<br />
BTS class (BTS > -272, Hoelzle 1992). The<br />
highest located point in “permafrost improbable’’ BTSclass<br />
was at 283 m a.s.1.; the highest point in “permafrost<br />
possible” BTS class was measured at 478 m, whereas several<br />
points in the “permafrost probable” BTS class were<br />
found near sea level. The results are supported by borehole<br />
temperatures from the area showing permafrost at<br />
87
Kristensen. 1,: Using BTS for mapping perniafrost distribution ...<br />
360 m a.s.1.<br />
The BTS measurements were analyzed with respect to<br />
altitude, PR and snow depth. Within the measured interval<br />
of snow depths, no significant relation between BTS and<br />
snow depth was found. Altitude explains 59 % of the variance<br />
in BTS and PR <strong>25</strong> YO. However, a significant negative<br />
correlation (r -0,33) between PR and BTS shows<br />
that the dataset is biased, indicating that areas facing<br />
south are more likely to have been sampled at low altitudes.<br />
Because of this multiple regression, using both altitude<br />
and PR as the controlling variables only improves<br />
R2 to 66 %. The relation takes the form:<br />
BTS = -0.0054~ altitude + 0,028 x PR - 5.30<br />
The equation was applied to the entire study area,<br />
simulating BTS throughout the DEM. 121 BTS measurements<br />
were assigned to the appropriate BTS class, 33<br />
were assigned to the neighbouring BTS class, whereas six<br />
were assigned to two BTS classes from the measured<br />
one. According to this relation, permafrost limits are highly<br />
influenced by radiation. Areas receiving low amounts of<br />
radiation are modelled to belong to the “permafrost probable”<br />
BTS class at sea level, whereas areas receiving<br />
maximum radiation are modelled to the “permafrost improbable”<br />
BTS class up to <strong>25</strong>0 m a.s.1. and “permafrost<br />
possible” up to 430 m a.s.1. The importance of altitude and<br />
PR in explaining BTS is comparable to findings from Jotunheimen<br />
and Dovrefjell, southern Norway (approximately<br />
62’), though permafrost in these areas is found at<br />
higher altitudes (Odegard 1996). Radiation is probably of<br />
much greater importance in the eastern Swiss Alps (approximately<br />
46”, Hoelzle 1992) than in Greenland, which<br />
might reflect the lower latitude and higher amounts of radiation<br />
received in the Alps.<br />
sults are supported by borehole temperatures from the<br />
area showing permafrost at 360 m a.s.1. However, multiple<br />
regression including both altitude and PR shows that the<br />
thermal regime of the ground is highly influenced by PR as<br />
permafrost-free areas or areas marginal to permafrost are<br />
found up to nearly 500 m a.s.1. where PR is high.<br />
REFERENCES<br />
Christiansen, H. H. and Humlum, 0. <strong>20</strong>00. Permafrost. In B. H.<br />
Jakobsen. Bocher, J. Nielsen, N. Guttesen, R. Humlum, 0. and<br />
Jensen, E (eds), Topografisk Altas Grernland. Kerbenhavn: C.A.<br />
Reitzels Foriag.<br />
Haeberli, W. 1973. Die Basis-Temperatur der winterlichen<br />
Schneedecke als moglicher indicator fur die Verbreitung von<br />
Permafrost in den Alpen. Zeitschrift fur Gletscherkunde und<br />
Glazialgeologie 9: 221 -227.<br />
Hoelzle, M. (1992). Permafrost occurrence from BTS measurements<br />
and climatic parameters in the Eastern Swiss Alps. Permafrost<br />
and Periglacial Processes 3: 143-147.<br />
Weidick, A. 1968. Observations on some Holocene glacier fluctuatuins<br />
in West Greenland. Meddelelser om Granland. 164(6): 1-<br />
<strong>20</strong>2.<br />
0degArd. R. S. Hoel.de, M. Johansen, K. V. and Sollid, J. L. 1996.<br />
Permafrost mapping and prospecting in southern Norway. Norsk<br />
geogr. Tidsskr. 50: 41-53.<br />
Figure 1. The effect of<br />
found by the model.<br />
PR on BTS classes and permafrost limits as<br />
CONCLUSIONS<br />
BTS measurements in the Apussuit region indicate widespread<br />
permafrost in the area, even at sea level. The re-<br />
88
Measuring rock glacier surface velocities with real time kinematics GPS<br />
(Mont Gele area, western Swiss Alps)<br />
C. lambiel, *R. Delaloye, **l. Baron and R. Monnet<br />
Institute of Geography, University of Lausanne, <strong>Switzerland</strong><br />
*Dept. of Geosciences, Geography, University of Fribourg, <strong>Switzerland</strong><br />
**/nsfitufe of Geophysics, University of Lausanne, <strong>Switzerland</strong><br />
INTRODUCTION<br />
Studies on rock glacier movement have been led for a<br />
long time on several sites in the Alps (e.9. Barsch 1992).<br />
Methods applied until now are essentially measurements<br />
using theodolites (ens. Francou and Reynaud 1992) and<br />
aerial photogrammetry (e.g. Kaab et al. 1997). Rock glacier<br />
surface velocities measurements with Real Time<br />
Kinematics (RTK) GPS have been tested in the Mont Gele<br />
area (Verbier, Swiss Alps, 46"06'N/7"17'E).<br />
The studies were conducted on three rock glaciers, two<br />
of them (B and C) north-northeastern oriented and the<br />
third one (D) facing towards the east. Rock glacier D is<br />
mainly inactive (presence of depressed parts, no steep<br />
front, abundance of lichens), whereas rock glaciers B and<br />
C are clearly active (very steep fronts, unstable blocks, no<br />
lichens). The roots of rock glacier C are largely covered by<br />
an ice patch. The occurrence of permafrost is corroborated<br />
in all rock glaciers by many BTS measurements<br />
performed since 1993 and by continuous logging of the<br />
subsurface temperature using miniature loggers UTL-1.<br />
In summer 1998, DC resistivity soundings and mapping<br />
lines were carried out on the rock glaciers (Reynard<br />
et al. 1999). Some of the results are presented in Figure 1.<br />
Marked differences in resistivities are primarily interpreted<br />
as differences in ice content. They also could reveal differences<br />
in temperature of the icehock mixture. According<br />
to Reynard et al. (1999)) rock glaciers B and D contain<br />
congelation ground ice and are therefore considered as<br />
typical periglacial rock glaciers. Low resistivities (max. 50<br />
kWm) should indicate a permafrost temperature close to<br />
the thawing point. Feature C is more complex. The presence<br />
of the ice patch at its roots and the high resistivities<br />
measured (specific resistivity of about 350 kWm) indicate<br />
the presence of massive (and colder) ice within the rock<br />
glacier. The resistivity mapping demonstrated that there<br />
was no break between the ice patch in the rooting zone<br />
and the rock glacier (Reynard et al. 1999).<br />
Figure 1. Three of the DC resistivity soundings carried out<br />
Mont Gel6 rock glaciers (Reynard et al. 1999, modified).<br />
METHOD<br />
on the<br />
The method used to survey the rock glacier surface velocities<br />
is of Real Time Kinematics type (Leica Geosystems):<br />
two stations (reference and rover) permits the local<br />
coordinates to be localized after a few seconds with a<br />
relative accuracy of 2-3 cm in the horizontal (x, y) coordinates<br />
and 3-5 cm in the vertical coordinate (z).<br />
In September <strong>20</strong>00, 87 blocks were measured on the<br />
rock glaciers. Four control points were also surveyed on<br />
bedrock, i.e. at places that are not affected by movement.<br />
Only blocks that seemed to be deeply buried in the active<br />
layer (and if possible in the permafrost body) were chosen.<br />
One year later, the position of the blocks was measured<br />
again.<br />
89
RESULTS AND INTERPRETATION<br />
Results are presented in Figure 2. A spectacular difference<br />
in horizontal movements between rock glaciers B<br />
and C can be observed. Rock glacier B clearly shows the<br />
maximal surface velocities. Annual displacement is up to<br />
1<strong>25</strong> cm a-1 in the central part of the rock glacier. Such values<br />
are relatively high compared to values usually obtained<br />
on rock glaciers (Barsch 1992). The lowest movements<br />
are encountered in the lateral parts of the<br />
formation. The roots show smaller velocities than the central<br />
and frontal parts of the rock glacier. Rock glacier C is<br />
also affected by movements, but to a much lesser extent.<br />
Maxima are encountered on the right side, toward the<br />
frontal part. The left side and the upper flat area appear to<br />
be more stable. Rock glacier D only shows velocities<br />
lower than 5 cm a-I.<br />
slope becomes steeper. This part corresponds to the<br />
faster velocities. As no flat zone is present further down,<br />
velocities remain high. Lower velocities measured on rock<br />
glacier C, despite higher resistivities, can be explained by<br />
topography as well. The slope of the formation is much<br />
less regular and large flat areas dominate the rock glacier.<br />
The presence of obstacles (rock glacier D and an outcrop<br />
of roches moutonnees) at the front of the rock glacier can<br />
also explain the relatively low velocities of this part. This<br />
forces the rock glacier to turn with an angle of about 90"<br />
to the right. Such a change in direction probably affects<br />
the velocity of the rock glacier.<br />
The second hypothesis refers to differences in icelrock<br />
mixture temperature. Rock glaciers with temperature close<br />
to the thawing point would creep faster than colder rock<br />
glaciers (e.g., lkeda et al. <strong>20</strong>03). Nevertheless, except for<br />
the difference in electrical resistivity, there is still no further<br />
data available confirming any difference in the temperatures<br />
of both rock glaciers B and C.<br />
CONCLUSION<br />
RTK GPS has proved to be very efficient for rock glacier<br />
surface velocity measurements. The precision of a few<br />
centimeters is sufficient and one (long) day of work permits<br />
more than a hundred points to be measured.<br />
This ongoing study has suggested that surface velocities<br />
of rock glaciers are not governed by ice content, but<br />
are more closely related to the topography (slope angle) of<br />
the rock glacier or to the temperature of the icelrock mixture.<br />
REFERENCES<br />
Horizontal surface velocities (cm a")<br />
A 100to 1<strong>25</strong> F 11 to<strong>25</strong> 0
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Detection of frost heave and thaw settlement in tundra<br />
environments: application of differential GPS technology<br />
J. Little, H. Sandall, M. Walegur and F. Nelson<br />
Department of Geography and Center for Climatic <strong>Research</strong>, University of Delaware, Newark, USA<br />
Geocryologists have devised several methods to measure<br />
the vertical component of frost heave and thaw subsidence.<br />
Most have employed either (a) mechanical<br />
“heavometers” that record differential movement at points<br />
or over very small (e.g. 10 m*) areas; or (b) optical leveling,<br />
used in conjunction with specially designed platform<br />
targets. Measurements are usually made within localized<br />
coordinate systems and have rarely been related precisely<br />
to geodetic coordinate systems.<br />
Degradation of permafrost and thaw settlement have<br />
received much attention recently in popular and globalchange<br />
literature (e.g. McCarthy et al. <strong>20</strong>01, Goldman<br />
<strong>20</strong>02). These processes pose considerable hazard to human<br />
activities and infrastructure in high latitude regions.<br />
Recent technological advances in geodetic science have<br />
the potential to measure vertical movements accurately<br />
and rapidly, to survey relatively large areas, and to communicate<br />
survey results in precise, externally referenced<br />
coordinate systems. Among the most promising technological<br />
developments are Differential Global Positioning<br />
Systems (DGPS).<br />
During summer <strong>20</strong>01 and <strong>20</strong>02, post-processed sfopand-go<br />
kinematic, rapid-static carrier phase DGPS, and a<br />
total of sixty-four specially designed heave and subsidence<br />
targets were used to record frost heave and ground<br />
subsidence at two 1 ha sites in northern Alaska. One site<br />
is at Sagwon in the northern foothills of the Brooks Range,<br />
the other at West Dock in the Prudhoe Bay oilfield on the<br />
Coastal Plain (Fig. 1). Our preliminary results indicate that<br />
post-processed rapid-static carrier phase DGPS has the<br />
ability to measure frost heave and ground subsidence at<br />
very high resolution (e.g. I cm).<br />
Global Positioning Systems acquire coordinate positions<br />
through triangulation of an antenna receiver with at<br />
least four satellites. DGPS, which requires two antenna<br />
receivers, improves the accuracy to the cm scale; we used<br />
a base receiver (Trimble 4000) and a rover receiver<br />
(Trimble 4700). The inexpensive platform targets are cylindrical<br />
platforms capable of supporting a rover DGPS<br />
antenna while allowing the freedom to move vertically in<br />
response to heave or settlement. The ability to move the<br />
antenna from target to target permits acquisition of data at<br />
time scales varying from 15 to 45 seconds for stop-and-go<br />
kinematic DGPS; rapid-static DGPS requires eight to<br />
twenty minutes, but improves accuracy substantially (Little<br />
et al. in review).<br />
The West Dock and Sagwon sites, located in the Kuparuk<br />
River basin of northern Alaska (Fig. I), were instrumented<br />
with 32 platform targets, arranged according to a<br />
nested hierarchical sampling design (Nelson et al. 1999).<br />
The West Dock site is composed primarily of wet nonacidic<br />
tundra, and is underlain by ice-wedge polygons.<br />
The study site occupies part of a drained thaw lake basin<br />
and an adjacent section of “upland” tundra, which has<br />
better drainage. The Sagwon site occupies a section of<br />
nonacidic tundra with frost boils and patchy vegetative<br />
cover (Walker et al. 1998).<br />
Figure 1. Map showing locations of West Dock and Sagwon (Flux 3)<br />
sites, North Slope, Alaska near the Dalton Highway.<br />
All targets were given an arbitrary position of 10 cm at<br />
the beginning of the survey during summer <strong>20</strong>01; Figure 2<br />
shows heave and subsidence relative to this position.<br />
Rapid-static DGPS observations for June <strong>20</strong>02 resulted in<br />
average surface heave of 1 cm from <strong>July</strong> <strong>20</strong>01 (Fig. 2a)<br />
with maximum heave and subsidence of 6.7 cm and 2 cm,<br />
respectively. Although observations were recorded in <strong>July</strong><br />
<strong>20</strong>01 instead of August, the normal period of maximum<br />
thaw in northern Alaska, 24 of 32 targets recorded heave<br />
by June <strong>20</strong>02. Mean values from a rapid-static DGPS survey<br />
in August <strong>20</strong>02 were 4.3 cm lower than in June <strong>20</strong>02.<br />
91
Subsidence was recorded at 29 of 32 targets, with maximum<br />
subsidence of 15.5 cm and heave of 0.5 cm (Fig.<br />
2a).<br />
Heave of 1.9 cm was recorded from August <strong>20</strong>01 to<br />
June <strong>20</strong>02. Of the 32 targets, 24 recorded heave in June<br />
<strong>20</strong>02 (Fig. 2b). Rapid-static DGPS from August <strong>20</strong>02 recorded<br />
3.3 cm of subsidence from June <strong>20</strong>02 (Fig. 2b).<br />
2a <strong>20</strong><br />
E<br />
0 T'2L<br />
-5 n<br />
-10<br />
June 02 August 02<br />
2b 30 7<br />
2c<br />
-In<br />
T<br />
June 02 August 02<br />
40 3 T<br />
classical surveying and rapid-static DGPS, respectively,<br />
yet were 14.2 for stop and go kinematic DGPS (Fig. 2c).<br />
The close correspondence between traditional survey<br />
methods and rapid-static DGPS at West Dock indicates<br />
that frost heave and thaw can be successfully detected<br />
with the latter. As part of an ongoing project, the vertical<br />
displacement of each platform will be measured again<br />
using DGPS in summer <strong>20</strong>03.<br />
ACKNOWLEDGEMENTS<br />
This DaDer is a contribution to th e CircumDolar Activ e<br />
Laye; Monitoring (CALM) program. We are grateful to<br />
Shad O'Neel (UNAVCO) for making DGPS equipment<br />
available for work in Alaska and for formal instruction in its<br />
use at UNAVCO's Boulder, CO facility. Professor Carmine<br />
Balascio (Department of Civil Engineering, University of<br />
Delaware) kindly provided surveying equipment for the<br />
validation work. We thank British Petroleum (Alaska) Inc.<br />
for logistical support and its Health, Safety and Environment<br />
office for access to the West Dock site. Funding was<br />
provided by the US. National Science Foundation under<br />
Grant OPP-0095088 to F.E. Nelson and Grant OPP-<br />
973<strong>20</strong>51 to K.M. Hinkel (University of Cincinnati). Any<br />
opinions, findings, conclusions, or recommendations expressed<br />
in this material are those of the authors and do<br />
not necessarily reflect the views of the National Science<br />
Foundation. Identification of specific products and manufacturers<br />
in the text does not imply endorsement by the<br />
National Science Foundation.<br />
Classical Rapid Static Stop and go kinematic<br />
Figure 2a. Box plot depicting heave June <strong>20</strong>02 from <strong>July</strong> <strong>20</strong>01 at<br />
West Dock. Subsidence occurred between June and August <strong>20</strong>02 and<br />
was detected by rapid-static DGPS. Figure 2b. Box plot showing<br />
heave and subsidence at Flux 3. Figure 2c. Box plot comparing results<br />
from traditional surveying with stop and go kinematic and rapidstatic<br />
DGPS.<br />
The use of DGPS as a stand-alone replacement for<br />
traditional survey methods (e.g., differential leveling or<br />
profile leveling) has been called into question (e.g.! US<br />
Army Corps of Engineers 1996). To ascertain the degree<br />
of correspondence between DGPS and traditional survey<br />
techniques in the context of frost heave and thaw settlement,<br />
we compared results at West Dock from rapid-static<br />
and stop and go kinematic DGPS with those from a survey<br />
using profile leveling. The traditional survey, conducted in<br />
August <strong>20</strong>02, detected mean subsidence of 2 cm, compared<br />
with results from June <strong>20</strong>02 (Fig. 2c). Comparison<br />
of results from traditional surveying and the two DGPS<br />
techniques confirmed that rapid-static DGPS yields much<br />
better accuracy. Profile leveling in June <strong>20</strong>02 averaged<br />
just 1.2 cm higher than rapid-static DEPS recorded a few<br />
days earlier, but 5 cm higher than stop-and-go kinematic<br />
(Fig. 2c). Standard deviations were 1.6 and 3.9 cm for<br />
REFERENCES<br />
Goldman, E. <strong>20</strong>02. Even in the High Arctic, Nothing is Permanent.<br />
Science, 297: 1493-1494.<br />
Little, J.D., Sandall H., Walegur, M.T., and Nelson, F.E in review. Application<br />
of differential GPS to monitor frost heave and ground<br />
subsidence in tundra environments. Permafrost and Periglacial<br />
Processes.<br />
McCarthy, J.J., Canziani, O.F., Leary, N.A., Dokken, D.J. and White,<br />
K.S. <strong>20</strong>01. Climate Change <strong>20</strong>01: Impacts, Adaptation, and Vulnerability.<br />
Cambridge University Press, Cambridge, UK.<br />
US Army Corps of Engineers. 1996. Engineering and Design -<br />
NAVSTAR Global Positioning System Surveying. April, <strong>20</strong>02.<br />
http:I/www.usace.army.miI/usace-docs/eng-manuals/em111O-1-<br />
1003/toc.htm.<br />
Walker, D.A., Auerbach, N., Bockheim, J.G., Chapin, F.S. 1.1.1,, Eugster,<br />
W., King, J.Y., McFadden, J.P., Michaelson, G.J., Nelson,<br />
F.E., Oechel, W.C., Ping, C.L., Reeburgh, W.S., Regli, S., Shiklomanov,<br />
N.I., and Vourlitis, G.L., 1998: A major arctic soil pH<br />
boundary: implications for energy and trace-gas fluxes. Nature<br />
394: 469-472.<br />
92
Permafrost and Little Ice Age glaciers relationships: a case study in the<br />
Posets Massif, Central Pyrenees, Spain<br />
R. lugon, *R. Delaloye, **E. Serrano, ***E. Reynard and C. lambiel<br />
University Institute Kurt Bosch, Sion, <strong>Switzerland</strong><br />
*Department of Geosciences, University of Fribourg, <strong>Switzerland</strong><br />
**Department of Geography, University of Valladolid, Spain<br />
***Institute of Geography, University of Lausanne, <strong>Switzerland</strong><br />
INTRODUCTION<br />
The prospecting of ground ice in sediment deposits within a<br />
glacier forefield requires two main types of ground ice to be<br />
differentiated: pure permafrost ice (congelation ice within the<br />
ground) and debris-covered (or buried) glacier ice. DC (direct<br />
current) electrical prospecting is an easy and light technique,<br />
which allows such a distinction to be revealed in many situations<br />
(Haeberli and Vonder Muhlll996).<br />
Resultsfromageoelectricalmeasuringcampaignin the<br />
forefield of the Posets glacier are briefly presented and discussed<br />
here. As far as we know, this is the first study treating<br />
the relationship glacier - permafrost (rock glacier) within<br />
the Little Ice Age (LIA) Pyrenean glacier forefields.<br />
Hummel configuration allowed detection of shifts in ground resistivity<br />
in both branches of the soundings.<br />
The VES were partially complemented with resistivity<br />
mapping (wenner configuration). The technique allowed spatial<br />
changes in the measured apparent resistivity (usually 2-<br />
10 times lower than the specific resistivity for a permafrost<br />
body) to be determined at a fixed pseudo-depth.<br />
A challenging aspect was to dispose of a sufficient electrical<br />
supply for a whole week of measuring. An OYO<br />
McOhm device (model 2115), complemented with a single<br />
lithium-lead battery (about 10 kg), was well suited as no recharge<br />
was required. This was only possible due to the use<br />
of low-intensity current (between 1 and 5 mA). Sponges<br />
soaked with salt water were used as electrodes.<br />
STUDY AREA<br />
The LIA margins of the Posets glacier are located between<br />
2900 and 3100 m a.s.1. on the eastern side of the Posets<br />
peak (3369 m a.s.1.) in the Central Pyrenees (42"39'N,<br />
0'26'E). In <strong>20</strong>01, the Posets glacier covered less than 0.1<br />
km2 (Fig. I). It was three to four times larger during the LIA.<br />
At that time, the glacier divided into two tongues, one flowing<br />
to the east towards the Posets rock glacier, and the other b<br />
the north towards the La Paul rock glacier. The presence d<br />
permafrost in the glacierhock glaciers complex was shown<br />
by Serranoet ai. (<strong>20</strong>01) who used a setof BTS (Bottom<br />
Temperature of the Snow cover) and year-round continuous<br />
ground temperature measurements.<br />
The bedrock beneath almost all the glacial sediment aG<br />
cumulations consists of a granite batholith. Hornfels resulting<br />
from contact metamorphism of Paleozoic shales surrounds<br />
the igneous intrusion and forms the greater part of the face an<br />
Posets peak overhanging the glacier cirque.<br />
PROSPECTING METHODS<br />
Two complementary DC resistivity techniques were applied.<br />
Vertical electrical soundings (VES) were used to estimate the<br />
stratigraphy of the ground resistivity. The dissymmetrical<br />
RESULTS AND INTERPRETATION<br />
Posefs rock glacier<br />
Two VES were performed on the Posets rock glacier, a 400<br />
m long imposing sedimentary formation located at the front d<br />
the former eastern tongue of the Posets glacier (Fig.1). Five<br />
main arched ridges are clearly visible. Under a thin (1 m) active<br />
layer (<strong>20</strong> kam), mainly consisting of angular homfel<br />
clasts, an extreme resistive layer was detected (specific re<br />
sistivity: 3-12 MQm, 5-12 m thick), beneath which the resistivity<br />
falls towards few a tens of kQm.<br />
The spatial extension of the apparent resistivity of the<br />
second layer was mapped at a depth of about 6-9 m (AB<br />
37.5 m, 78 measurements). Extreme apparent resistivities,<br />
somewhere higher than 500 kQm, were measured in the<br />
whole sedimentary complex. However, resistivities strongly<br />
decreased towards the rooting zone of the debris rock glacier<br />
where values between 50 and I00 kQm were found.<br />
The high resistivity of the second layer clearly indicates<br />
the presence of a massive ice layer, immediately beneath a<br />
thin layer of deposits. Such massive ice is interpreted as<br />
buried ice of glacial origin. The very high resistivity of the<br />
second layer could have madedeeper layer(s) with lower<br />
resistivities undetectable, i.e., permafrost with low resistive<br />
ground ice.<br />
93
la Paul rock glacier<br />
The 400 mlongLa Paul rock glacier shows a typical but<br />
poorly-developed relief of permahst creep with transversal<br />
furrows and ridges. A VES revealed the presence of an extremely<br />
resistive layer (8-<strong>25</strong> MQm, 5-15 m thick) below a 2<br />
m thick active layer (3040 kQm). The downward branch d<br />
the VES shows the decreased resistivity of the second layer<br />
towards the frontal part of the rock glacier (1.3-5 MQm, 515<br />
m thick). As for the Posets rock glacier, geoelectrical methods<br />
could not prove nor exclude the existence of a third underlying<br />
frozen layer.<br />
The extreme values of resistivity calculated towards the<br />
rooting zones of the rock glacier are only known to be found<br />
in cold or polythermal glaciers (buried sedimentary ice, Haeberli<br />
and Vonder Muhlll996).<br />
FURTHER QUESTIONS<br />
Geo electrical measurements pointed to extreme resistivity<br />
ice in two debris rock glaciers that were prospected. Specific<br />
resitivities ranging between 1 and <strong>25</strong> MQm were detected<br />
under a thin (1-2 m) active layer. The following suppositions<br />
can be made about the origin of ice within both Posets La and<br />
Paul rock glaciers.<br />
lugon, R. et 81.: Permafrost and Little Ice Age glacicrs ...<br />
Debris-covered glacier ice could have been incorporated<br />
in the two rock glaciers during the advance of the Posets glacier<br />
in the LIA, or former Holocene advances. Very cold<br />
subsurface temperatures measured in the deposits (Serrano<br />
et al. <strong>20</strong>01) indicate that buried ice could easily have been<br />
preserved since the end of the LIA.<br />
The massive ice probablysuperimposedformerpermafrost<br />
bodies, i.e., a much thicker layer of perennially frozen<br />
ground. However, such permafrost could not be detected by<br />
means of geo-electrical techniques.<br />
The processes which have created both La Paul and<br />
Posets debris rock glaciers therefore remain in question.<br />
What is the role played by the glacier dynamic in the<br />
formation of the deposits What is the importance of the<br />
creeping processes Do the ridges occurring at the surface d<br />
the Posets glacier attest to advances of the Posets glacier<br />
during LIA, or do they indicate a sign of permafrost creeping<br />
REFERENCES<br />
Haeberli, W. and Vonder Muhll, D. 1996. On the characteristics and<br />
possible origins of ice in rock glacier permafrost. Zeitschrift fur<br />
Geomorphologie, Suppl. Bd. 104: 43-57.<br />
Serrano, E., Agudo, C., Delaloye, R. and Gonzalez-Trueba, J.J.<br />
<strong>20</strong>01. Permafrost distribution in the Posets massif, Central<br />
Pyrenees, Norsk Geografisk TidsskrifbNorwegian Journal of<br />
Geography 55: 245-<strong>25</strong>2.<br />
Little Ice Age Pnsets ~ l ~ ~ i e ~<br />
94
Borehole drilling in permafrost of an avalanche-affected slope at<br />
Fluelapass, eastern Swiss Alps<br />
M. luetschg, V. Sfoecki, W. Ammann and *W. Haeberli<br />
Swiss Federal institute for Snow and Avalanche <strong>Research</strong> (SLF), Davos Doe <strong>Switzerland</strong><br />
*Glaciology and Geomorphodynamics Group, Department of Geography, University of <strong>Zurich</strong>, <strong>Zurich</strong>, <strong>Switzerland</strong><br />
Alpine permafrost is affected by different topographic and<br />
climatic factors. The snow cover is of high importance due<br />
to its strong insulation and albedo effects. Snow cover has<br />
a warming effect on ground temperatures during cold<br />
winter, whereas in late spring and summer, snow cover<br />
has a cooling effect (Keller 1994). This means that not<br />
only altitude and aspect but also snow-cover distribution is<br />
important with regard to the permafrost distribution; avalanches,<br />
therefore, play an important role in permafrost<br />
distribution, as they redistribute, compress and accumulate<br />
the snow: hence, at the accumulation zone of an<br />
avalanche slope, cooler ground temperatures were measured<br />
than in the passing zone higher above, due to the<br />
long remaining avalanche snow (Haeberli 1975). Thus,<br />
permafrost often occurs in the lower parts of avalanche<br />
slopes.<br />
Such an avalanche-affected permafrost slope was<br />
found at Fluelapass, located in the eastern Swiss Alps, at<br />
an altitude of 2380 m a. s. I. The slope above the lake<br />
Schottensee is exposed towards northeast and is between<br />
33 and 37 degrees steep at the upper part of the slope,<br />
corresponding to a typical avalanche starting zone.<br />
Avalanches indeed occur in this slope in late winter<br />
and spring and deposit the snow at the foot of the slope.<br />
These large snow accumulations disappear later in the<br />
season than the snow on the slope above. This fact was<br />
also observed by a series of automatically-taken photographs<br />
during I999 (Lerjen et al. <strong>20</strong>03). Measurements of<br />
ground surface temperatures and seismic refraction<br />
soundings indicate colder temperatures and the presence<br />
of ice close to the surface in the lower third of the slope<br />
(Haeberli 1975). In summer 1999, these measurements<br />
were repeated and the active layer thickness was determined<br />
to be between 2 and 4 m in the lower part of the<br />
slope (Lerjen et al. <strong>20</strong>03).<br />
Ground surface temperatures in the slope and at both<br />
shores of the lake were measured during winter <strong>20</strong>01102.<br />
In summer <strong>20</strong>02, two boreholes were drilled, one (bh I)<br />
in the avalanche starting zone (7913521180353 <strong>25</strong>00 m a.<br />
s. I.) and one (bh 2) at the foot of the slope<br />
(7915001180474 2394 m a. s. I.) to study the effect from<br />
the redistribution of snow due to avalanche events on<br />
ground temperatures. In the lower borehole (bh 2), cores<br />
containing pore ice were sporadically drilled. Figure l-A<br />
shows the stratigraphy of the two boreholes as based on<br />
information from the experienced borehole-drilling team,<br />
and Figure l-B presents the temperature profiles which<br />
were measured at the beginning of winter <strong>20</strong>02103, six<br />
weeks after the end of the drilling.<br />
In the upper borehole (bh I), 40 cm of humus to sandy<br />
soil was found under a dense vegetation cover (Fig. 1-A).<br />
This top layer was underlain by blocky material which<br />
contained larger blocks in the lower part. At 14.5 m depth,<br />
bedrock was encountered. The lower borehole (bh 2) differs<br />
from the upper borehole by the lack of a vegetation<br />
cover and the presence of pore ice: this ice was found<br />
between 1.8 and 8.5 m depth. Cores were drilled at the<br />
depths from 3.3 to 4.<strong>25</strong> m and from 5.1 to 6.44 m; they<br />
contained some fragments of the pore ice. In the lower<br />
borehole, bedrock was not reached but at <strong>20</strong> m depth, the<br />
porous material was filled with water, indicating that the<br />
level of the lake surface was reached.<br />
In both boreholes, inclinometer tubes were inserted to<br />
measure future deformation of the scree slope. Additional<br />
surveyings of the borehole positions (surface coordinates)<br />
are performed twice a year. Ground temperatures have<br />
been monitored continuously since October <strong>20</strong>02.<br />
First borehole temperature measurements at the beginning<br />
of winter <strong>20</strong>02103 reveal a clear difference between<br />
the two positions in the slope. In the lower borehole<br />
(bh 2), temperatures are close to 0°C (fig. 1-B), while<br />
below 13 m, the temperature in borehole bh 2 increases to<br />
attain positive values. Note that the spatial resolution of<br />
the temperature profile in the lower part is 5 m, corresponding<br />
to the distance of the installed temperature sensors.<br />
Since winter <strong>20</strong>01102, surface temperatures at four positions<br />
between the two borehole positions were measured<br />
continuously with universal temperature loggers<br />
(UTL) (Fig. 2). They show that the UTLs placed at lower<br />
altitudes close to the lower borehole position are becoming<br />
snow-free about 80 days later than in the upper part of<br />
95
the slope. BTS were measured at the end of April <strong>20</strong>02 to<br />
be - 2.8 "C in the upper part and - 3.8 "C in the lower<br />
part of the slope.<br />
The borehole temperatures are supposed to still be<br />
somewhat affected by the effects of drilling disturbance,<br />
due to the short period between the end of drilling and the<br />
first temperature measurements presented in Figure 1-B.<br />
Results from the studies presented here will be used in<br />
comparison to modelling results, in which the situation is<br />
calculated with the one-dimensional model SNOWPACK<br />
(Bartelt and Lehning <strong>20</strong>02). This snow-cover model was<br />
developed and extended to the underlying ground. It will<br />
be used to study the effect of snow height and snow density<br />
on ground temperatures in the avalanche starting<br />
zone and at the foot of the slope.<br />
Borehole temperature measurements, surveys of snow<br />
cover characteristics and snow distribution and numerical<br />
simulations will give a better understanding of the effect of<br />
avalanches on permafrost distribution in avalanche<br />
slopes.<br />
T["CI<br />
bh 1 bh 2 P l ! i l I<br />
ACKNOWLEDGEMENTS<br />
The borehole drilling presented here was carried out by<br />
Stump Bohr AG. This work was partly funded by the Swiss<br />
National Science Foundation (Nr. 21-65263.01).<br />
REFERENCES<br />
Haeberli, W. 1975. Untersuchungen zur Verbreitung von Permafrost<br />
zwischen Fluelapass und Pit Grialetsch (Graubunden). Mitteilung<br />
der Versuchsanstalt fur Wasserbau, Hydrologie und Glaziologie<br />
der ETH <strong>Zurich</strong> 17: 221.<br />
Keller, F. 1994. lnteraktion zwischen Schnee und Permafrost. Mitteilung<br />
der Versuchsanstalt fur Wasserbau, Hydrologie und Glatiologie<br />
der ETH <strong>Zurich</strong> 127: 145.<br />
Bartelt, P. and Lehning, M. <strong>20</strong>02. A physical SNOWPACK model for<br />
the Swiss avalanche warning. Part I: numerical model. Cold Regions<br />
Science and Technology 35(3): 123-145.<br />
Lerjen, M., Kaab, A., Hoelzle, M. and Haeberli, W. <strong>20</strong>03. Local distribution<br />
pattern of discontinuous mountain permafrost. A process<br />
study at Fluela Pass, Swiss Alps. 8th International Conference on<br />
Permafrost, <strong>Zurich</strong>, <strong>Switzerland</strong>.<br />
blocks, stoner,<br />
Ice<br />
blocks. stones<br />
block<br />
stnnm<br />
mo-e<br />
block<br />
sohd rock<br />
blocks.<br />
stoncs.<br />
blocks.<br />
stoner.<br />
porous<br />
blocks<br />
mome<br />
block. stones,<br />
WatB<br />
I<br />
I<br />
I<br />
I<br />
A<br />
B<br />
Figure 1. Stratigraphy of the the upper borehole (bh 1) and the lower<br />
borehole (bh 2) drilled <strong>20</strong>02 in an avalanche slope at Fluelapass (A),<br />
and the corresponding ground-temperature profiles T with depth z,<br />
measured six weeks after drilling (B).<br />
30 , I<br />
Figure 2. Ground surface temperatures (Ts) measured in <strong>20</strong>01/02<br />
between the two borehole positions in the avalanche slope at Fluelapass<br />
show a temperature inversion with altitude.<br />
96
<strong>Research</strong> on the relationship between wave velocity and the drillability of<br />
frozen clay<br />
Q. Y. Ma, *W. Ma, **M. F. Cai and ***Z. H. Zhang<br />
Anhui University of Science and Technology, Huainan Anhui 23<strong>20</strong>01, China<br />
*University of Science and Technology of Bejing, Beijng 700083, China<br />
"Sfate Key Laboratory of Frozen Soil Engineering, Cold and Arid Regions Environmental and Engineering <strong>Research</strong>ing<br />
Institute, CAS, Lanzhou 730000, China<br />
**University of Science and Technology of Bejing, Beijng 100083, China<br />
***Anhui Univemty of Science and Technology, Huainan Anhui 23<strong>20</strong>01, China<br />
From studies on the theory of elasticity we know that the wave. A specially made constant temperature device for b<br />
propagation of an acoustic wave in rock or frozen ground is<br />
closely related to the character of the medium. Once the lon-<br />
Zen clay and an axial pressure coupled measuring shelf<br />
were used to do the experiment. Measurements were made<br />
gitudinal wave velocity and the transverse (shear) wave to a precision of kO.1 "C at six temperatures of: -5"C, -7"C,<br />
velocity have been determined, the corresponding character- -lO°C,-12"C,<br />
-15Co,,and -17°C respectively. The experiistics<br />
of elastic deformation are also defined. The propagation mental results are shown in Table 1.<br />
velocity of the acoustic wave, as measured in frozen ground,<br />
reflects the likely drilling properties of this frozen ground. The<br />
propagation velocity of frozen ground represents some typical<br />
Table 1, The experimental results of longitudinal wave<br />
velocity and transverse wave velocity of frozen clays<br />
physical properties. Using this basic principle the relationship<br />
name temperature longitudinal wave transverse wave<br />
between wave velocity and the drillability of frozen ground<br />
"C velocity, mls velocity, mls<br />
was derived. The measurement of wave velocity has certain clay C1 -5 2363 1001<br />
advantageous characteristics. This test is quick to perform clay ~2 -7 2477 1077<br />
and not intrusive, hence it causes no disturbance to the clay C3 -10 <strong>25</strong>94 1170<br />
ground. Using the wave velocity to determine the drillability clay C4 -12 2653 1224<br />
of frozen ground is of great practical value.<br />
clay C5 -15 27<strong>25</strong> 1279<br />
clay C6 -17 2749 1300<br />
A segment of frozen clay at a depth of 219.93-224.82m in<br />
a mine shaft in the Anhui province of China was chosen. Tne Measuring the drillability of frozen clay requires that the<br />
density of clay was 17lOkglm3 and had a water content d crushing instrument or chisel should follow an identical course<br />
21.42%.<br />
to the actual drilling hole so that the measured indexes of the<br />
In the experiment, a UVM-2 acoustic wave instrument drillability do not vary with time, position or person, this leads<br />
was used. This instrument can, via the sing-around method, to an objective and scientific index. Besides, using the chisperform<br />
a delayed determination so that it is not affected by eling work per unit volume to reflect the drillability of frozen<br />
multiple reflection waves and the velocity of the acoustic clay makes it very comparable.Theimpact is easily conwave<br />
in the material can be measured rapidly and accu- trolled and sufficient load may be applied, even from a portrately.<br />
The method used is as follows: the sensor is covered able instrument, to crush frozen clay of any typical hardness.<br />
with petroleum grease and placed on both sides of the sam- Refemng to the research on the available experimental<br />
ple. The receiver position is adjusted while observing the sig- equipment for drilling in rock, a chiselling instrument made by<br />
nal through an oscilloscope. The location and width of the os- East North University has been chosen.<br />
cilloscope display should be adjusted so that the whole The experimental method is as follows.<br />
ultrasonic pulse can be seen in the window and an optimal<br />
(I) The moisture content and particle size distribution should<br />
wave form is attained. The measured propagation time is the be measured before the experiment.<br />
measured acoustic cycle minus both the sensor delay and (2) The sample should be fixed to the test frame within the<br />
the fixed delay. The acoustic velocity can be calculated on low-temperature room.<br />
this basis.<br />
(3) A disc, graduated into 24 equal parts, with a central<br />
The diameter of the experimental sample was 61mm, the hole of 60mm diameter was fixed to the sample.<br />
height of both the longitudinal and transverse wave samples<br />
was 50.7mm. The experimental system had a fixed delay d<br />
2.99s for the longitudinal wave and 2.00s for the transverse
tus. One person needs to ensure that the angle setting<br />
knob does not move while the other operates the hammer.<br />
It must be ensured that the hammer can drop<br />
freely without any restriction.<br />
(5) The angle setting knob was tumed 15" after each impact.<br />
The pieces of frozen clay that fell to the bottom d<br />
the hole were cleaned out after each five revolutions. In<br />
addition, the hole depth was measured after 240 im-<br />
pacts (IO revolutions) and 480 impacts (<strong>20</strong> revolutions).<br />
(6) The hole depth was measured with special vernier calipers<br />
on the left edge, in the middle, and on the tight<br />
edge of the hole, with a precision of 0.1 mm. Assuming<br />
the frozen clay surface was smooth, the measured average<br />
value was regarded as the hole depth.<br />
(7)The chiseling work per unit volume can be calculated<br />
with the chiseling depth as follows:<br />
A<br />
ne"=-<br />
V<br />
Ma! Q.Y, el al.: <strong>Research</strong> on the relationship between wave velocity ...<br />
40NA,<br />
d2H<br />
Where a is the chiseling work per unit volume, kgmlcm3;<br />
is the number of cycles, its actual value is 480; AO is the<br />
work from each impact with an actual value of 3.9kg.m; d is<br />
the diameter of the drilled hole, the actual diameter is lmm<br />
greater than that of the drill bit so, value d is 4.1 cm; H is the<br />
depth of the drilled hole in rnm and V is the volume in cm3.<br />
The experimental results are shown in Table 2.<br />
Table 2. Drilling depth and chiseling work per unit volume of frozen<br />
Ma, Q. Y., Peng, W. W. and fhu, Y. L. <strong>20</strong>02. Relationship<br />
clays<br />
Temperature, drilling depth, chiseling work per unit volume, artificially frozen clay. Chinese Journal<br />
c",, mm kgo@/cm3<br />
Engineering 21 (2): 290-294.<br />
-5 122.<strong>20</strong> 11.90<br />
Zhang, Z. H. and Ma, Q. Y. <strong>20</strong>01. <strong>Research</strong> present situation<br />
-7 116.57 12.48<br />
analysis on classification of rock drillability, Chinese<br />
-10 86.27 16.86<br />
Coal Science and Engineering 7(1): 39-45.<br />
-12 74.27 19.58<br />
-15 67.17 21.65<br />
-17 57.50 <strong>25</strong>.30<br />
It can be seen from the data in Table 2 that the longitudinal<br />
wave velocity, the transverse wave velocity and the chiseling<br />
work per unit volume of frozen clay all appear to display<br />
a tendency to increase as the temperature decreases. The<br />
equation relating the longitudinal wave velocity, the transverse<br />
wave velocity and the chiseling work per unit volume<br />
can be established according to the principle of regression<br />
analysis.<br />
The regression equations are expressed as follows:<br />
a= 0.807"~104~p2- 0.379cp -t 457.66, R~0.988<br />
Where: cp= longitudinal wave velocity, mls.<br />
a= 1.171"~10~~*~0.227~~ + 121.33, R=0.990<br />
Where: cs= transverse wave velocity, mls<br />
The multi-variable regression equation for longitudinal and<br />
transverse wave velocity and chiseling work per unit volume<br />
is expressed as follows:<br />
a = -0.158~~ + 0.244~~ + 139.849, R=0.973, F= 54.507<br />
The value of Fis 54.507 and can be derived. It is known<br />
that F0.01(2,3)=30.8, F0.005(2,3)=49.8 and F0.001(2,3)=148.5<br />
from the f distribution table. From this we get f > f0.005(2,3),<br />
which signifies that the regression effect is excellent and that<br />
the result of the multi-variable linear regression is acceptable.<br />
N<br />
From the results of the experiment we know that the longitudinal<br />
and transverse wave velocity and the chiseling<br />
work per unit volume of frozen clay all tend to increase as the<br />
temperature decreases. While the longitudinal and transverse<br />
wave velocity is higher and the frozen clay is harder, the<br />
drilling is more difficult; conversely, when the longitudinal and<br />
transverse wave velocity is lower, the drilling is relatively<br />
easy. The relevant equation between the longitudinal and<br />
transverse wave velocity and the drillability of frozen clay<br />
can be established fairly easily by adopting the method d<br />
multi-variable regression analysis. It indicates that using the<br />
longitudinal and transverse wave velocity to evaluate the<br />
drillability of frozen clay is feasible.<br />
98<br />
ACKNOWLEDGEMENTS<br />
The authors wish to express their gratitude to The Outstanding<br />
Youth Foundation for Scientific and Technological <strong>Research</strong> d<br />
Anhui Province (<strong>20</strong>01-28), State Key LaboratoryofFrozen<br />
Soil Engineering and Education Department of Anhui Province<br />
of China for their financial support.<br />
REFERENCES<br />
Peng, W. W. and Yuan, Y. Q. <strong>20</strong>00. Developing of a broadband ultrasonic<br />
transducer of frozen soils. Journal of Glaciology and<br />
Geocryology 22 (I): 85-89.<br />
among<br />
longitudinal and transverse wave velocities and temperature of<br />
of Rock Mechanics and<br />
and<br />
journal of
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available i ~ f<br />
Permafrost active layer as the concentrator of contaminants in northern<br />
regions<br />
V. N. Makarov<br />
Permafrost lnsfifufe SB RAS, Yakufsk, Russia<br />
Soils are said to be the “self-purification filters” of nature.<br />
However, they lose their decontaminating properties in<br />
permafrost zones. The reasons are: thinness of the soil<br />
profile, inadequate drainage, thermohydrogeochemical<br />
barrier to the migration of contaminants, annual freezing<br />
and consequent concentration of soil water contaminants,<br />
low geochemical and lowered chemical activity of soils<br />
(because of low temperatures), short biological life of soils<br />
during the year. These conditions cause increasing soil<br />
pollution under the pressure from industrial zones.<br />
Regional permafrost aquicludes at shallow depth from<br />
the ground surface and low negative temperatures cause<br />
a specific geochemical situation in the northern territories<br />
and a widespread development of cryogenic metamorphoses<br />
in underground waters. The common direction of<br />
this process is to rise mineralization; the amounts of chloride<br />
and sulfate increase in the waters and heavy metals<br />
accumulate. During repeated freeze-thaw cycles the majority<br />
of dissolved contaminants become concentrated in<br />
gravitational waters while forming cryopegs. Cryopeg formation<br />
is due to the intensity and duration of industrial activity<br />
and results in its localization in the older parts of<br />
populated areas (Makarov, Venzke <strong>20</strong>00).<br />
Studies of soil solution behavior indicate their functional<br />
peculiarities during annual cycles. The most contrasting<br />
changes of macro- and microelement fluxes, pH value<br />
and oxidation-reduction potential (these determine migration<br />
in the geochemical environments of the upper horizons<br />
of seasonally thawed layers) are observed before<br />
and after the cold period when sharp temperature changes<br />
and moisture phase transitions occur. Before the beginning<br />
of the cold period, the upper horizons of seasonally<br />
thawed layers accumulate most of the components<br />
that come from industrial geochemical fields. Cryogenic<br />
concentration and the migration of soluble contaminants<br />
deep into seasonally thawed layers, induced by freezing<br />
front action, results in a depletion of the high horizons in<br />
seasonally thawed layers during winter. It causes a reduction<br />
in the concentrations of contaminant components by<br />
the beginning of the warm period. Maximum seasonal fluctuations<br />
in the intensity of geochemical fields is typical for<br />
components such as Cr, Ag, Pb, Mo, S, Ca, Na, CI, Mg,<br />
Sn whose concentration in pore water changes by 3 to 30<br />
times (Fig. 1).<br />
f” /<br />
Figure 1, Seasonal variations in the concentration of chemical components<br />
in soil’s pore water in Yakutsk<br />
The penetration of anomalous industrial components to<br />
the top of permafrost depends on the concentration gradient<br />
and the geochemical properties of the chemical elements<br />
forming the anomaly.<br />
Because of the similarity in some chemical properties<br />
and ion radii, iron and manganese hydroxides can sorb<br />
heavy metals from a suspension and co-precipitate with<br />
other metals (Ponnamperuma 1969). Heavy metal retention<br />
by iron and manganese hydroxides increases with<br />
time due to metal diffusion into the solid phase. During<br />
periods of thawing soils become hydromorphic. This promotes<br />
the formation of a large number of amorphous iron<br />
hydroxides which increases the maximum adsorption of<br />
heavy metals (Ostroumov 1998). However, the presence<br />
of organic matter, even in small amounts, retards the<br />
crystallization of iron hydroxides, thus causing increased<br />
metal mobility. The highly mobile components of organic<br />
matter in frost-affected soils form stable organometallic<br />
components which have a high ability to migrate. Metal<br />
mobility in organic soils is strongly influenced by microbial<br />
metabolism. Thus, metal mobility in frost-affected organic
tion of three soil components, iron components, organic<br />
matter and microbial communities.<br />
The complicated combination of simultaneously occurring<br />
processes determines the content ratio of elements in<br />
the solid and liquid phases of industrial geochemical<br />
fields. We can summarize the effects of the processes<br />
using the thickness ratio of the chemical elements aureole<br />
zone in the solid and liquid phases. Absorption of the solid<br />
phase is predominant for some elements and dissolution<br />
in water predominant for others.<br />
The majority of contaminant solid phases is accumulated<br />
in the upper parts of the ground at depths of 15-<strong>20</strong><br />
cm. The depth of contaminant penetration can reach 2.5-<br />
2.7 m, including the top of the permafrost, depending on<br />
the source of industrial contaminants and the intensity of<br />
the cryogenesis processes (Makarov 1986). It is caused<br />
by the transition of components in soil solutions and<br />
ground waters and by further sedimentation on various<br />
sorbents (Fig. 2).<br />
Makarw, V.N.: Permafrost active layer as the concentrator ...<br />
Figure 2. Distribution of chemical components in the active layer and<br />
permafrost. 1 - anthropogenically affected soils; 2 - sand; 3 - sandy<br />
silt; 4 - clay; 5 - permafrost.<br />
The thickest and ecologically<br />
most dangerous con-<br />
taminant anomalies - Hg, Pb, P, s, Ag, Sn, Cu, Y, Zn, Cr<br />
are forming at the thermohydrogeochemical barrier in urban<br />
zones.<br />
REFERENCES<br />
Makarov, V.N. 1986. Geochemistry of frost-affected soils in urban areas.<br />
In: Geochemistry of Technogenesis, Novosibirsk, Nauka:<br />
118-124.<br />
Makarov, V.N., Venzke, J.-F. <strong>20</strong>00. Umweltbelastung und Permafrost<br />
in den Stadt- und Agrarokosystemen in Zentral-Jakutien, Sibirien.<br />
Geographische Rundschau 12, Braunschweig: 21-26.<br />
Ostroumov, V.E. 1998. Ion redistribution in soils upon freezing.<br />
Pochvovedenie 5: 614-619.<br />
Ponnamperuma F. N., Loy, T. A. and Tianco, E. M. 1969. Redox<br />
equilibria in flooded soils: The Manganese oxide systems. Soil<br />
Science, 108, 1 : 48-57.<br />
100
Borehole and active-layer monitoring in the northen Tien Shan<br />
(Kazakhstan)<br />
S. S. Marchenko<br />
Permafrost Institute, Russian Academy of Sciences, Kazakhstan Alpine Geocryological laboratory, Almaty, Kazakhstan<br />
The Zhusalykezen Mountain pass (43"05'N, 76"55'E,<br />
3330 m a.s.1.) is located within the discontinuous permafrost<br />
area of the Transili Alatau Range (frontal range of<br />
northern Tien Shan). The location is close to the city of<br />
Almaty, and therefore offers an opportunity for geocryological<br />
and ecological investigations of the natural system<br />
and systems affected by human activities.<br />
The Upper Pleistocene and Holocene moraines from<br />
the mountain descend into the nearby valleys. Vegetation<br />
cover is dominated by Kobresia. Mean annual, January<br />
and <strong>July</strong> air temperatures at the elevation 3300 m a.s.1.<br />
are -3.5"C, -13.7"C and 6.7"C, respectively. The maximum<br />
annual precipitation is about 1000 to 1100 mm. The<br />
snow cover develops in October. By the beginning of<br />
June, the thickness of snow cover varies from 1 to 5 cm<br />
up to I00 to 1<strong>20</strong> cm, relative to topography and wind activity.<br />
Depending on geomorphology and microclimatic conditions,<br />
the thickness of permafrost varies from IOto 15 m<br />
to 80 to 90 m on the northern slopes and is absent on the<br />
southern slopes. Temperatures of permafrost vary from<br />
-0.1"C to -0.45"C. There are 10 thermometric boreholes<br />
located on different slopes and aspects within an area of 3<br />
km2. Active layer thickness is determined by temperature<br />
measurements made in boreholes from 3 m up to 70 m in<br />
depth. Several CALM sites are located near a mountain<br />
pass. Initial results of the Circumpolar Active Layer Monitoring<br />
were reported previously (Brown et al. <strong>20</strong>00).<br />
Ground temperatures up to <strong>25</strong> m in depth have been<br />
measured since 1973 (Gorbunov and Nemov 1978). Both<br />
interpolation of temperature measurements and excavations<br />
in the moraines revealed active-layer thickness in the<br />
late summer for 1973 and 1974 (Table. I). The table includes<br />
mean active-layer thickness during 1974-1977 for<br />
Cl(K0) and C2(K1) sites. The average increase in mean<br />
annual air temperature for the last I00 years in the central<br />
part of Transili Alatau Range has been 0.02"Clyr. The average<br />
ten-year temperature for 1991-<strong>20</strong>00 has increased<br />
by 0.3"C in comparison with 1981-90. The greatest increase<br />
of temperature for the same period has been mean<br />
winter (0.4"C), maximum winter (0.9T) and minimum<br />
summer temperature (0.YC). The warmest years for the<br />
last decade of the <strong>20</strong>th century were 1997, 1998 and 1999,<br />
when<br />
Table 1. Active-layer thickness near<br />
1973-1977.<br />
fusalykezen Mountain pass for<br />
Excavations Altitude Depth of pits Active layer<br />
and boreholes thickness<br />
m a.s.1. m m<br />
Pits<br />
No 1<br />
No 2<br />
NO 4<br />
No 5<br />
No 10<br />
No 12<br />
Boreholes<br />
C1 (KO)<br />
C2 (KI)<br />
C3a<br />
c7<br />
c11<br />
3337<br />
3334<br />
3334<br />
3337<br />
3336<br />
3334<br />
3337<br />
3328<br />
3337<br />
3334<br />
3335<br />
6.0<br />
4.0<br />
4.3<br />
9.3<br />
9.0<br />
4.2<br />
<strong>25</strong>.0<br />
14.0<br />
10.3<br />
11.7<br />
13.0<br />
4.5<br />
4.0<br />
3.5<br />
4.0<br />
3.8<br />
3.7<br />
3.2<br />
3.4<br />
4.3<br />
3.5<br />
3.7<br />
mean annual air temperature was higher than the longterm<br />
average temperature by 1 .I"C, 1.49"C and 0.93"C,<br />
respectively. Analysis of meteorological conditions for<br />
summer time has shown that the summer temperatures<br />
and precipitation figures are currently rising (Fig. I).<br />
E 400<br />
300<br />
e<br />
<strong>20</strong>0<br />
100<br />
1930 1940 1950 1960 1970 1980 1990 <strong>20</strong>00<br />
Figure 1. Air temperatures (a) and precipitation (b) for the summer<br />
time (Central Transili Alatau Range, <strong>25</strong>00 m a.s.1.). 1 - annual, 2 -<br />
five-year average, 3 - linear trend.<br />
101
Thus, climatic processes during the <strong>20</strong>th century and<br />
especially during the last two decades have a great influence<br />
on the contemporary thermal state of permafrost.<br />
Geothermal observations indicate permafrost warming in<br />
the northern Tien Shan for the last 30 years. The permafrost<br />
temperatures increased during 1973 - <strong>20</strong>02 by 0.2 to<br />
0.3"C for undisturbed systems, and up to 0.6"C of those<br />
affected by human activities.<br />
In accordance with interpolation of borehole temperature<br />
data, active-layer thickness showed a significant increase<br />
during the last 30 years from values of 3.2 to 3.4 m<br />
in the 1970s to a maximum of 5.2 m.<br />
The average active-layer thickness for all measured<br />
sites increased by 23% in comparison with the early Seventies.<br />
But compared with the thawed depths for the new<br />
CALM sites (KO, K1, C7) the increase is much greater at<br />
42%. For example, at the CALM K1 site during 1974-<br />
1977, the average active-layer thickness was 3.4 m and<br />
maximum value of 3.5 m. During 1990-<strong>20</strong>00 at the same<br />
site the average active-layer thickness was 4.85 m with a<br />
maximum value of 5.2 m. At the CALM site KO, the average<br />
and maximum active-layer thickness were 3.2 m and<br />
3.3 m for 1974-1977 and 4.8 m and 5.1 m during 1990-<br />
<strong>20</strong>00.<br />
As mentioned earlier, 1998 was the warmest year on<br />
record in Transili Alatau since 1964. As a result, in the<br />
following year the ground temperature at a depth of 4.6 m<br />
had increased significantly. For example, the average<br />
temperatures at 4.6 m were -0.2,2.24,0.10,0.06 and 0.04<br />
in 1998, 1999, <strong>20</strong>00, <strong>20</strong>01, and <strong>20</strong>02, respectively. This<br />
deep penetration of seasonal thawing into the ground initiated<br />
the appearance of a residual thaw layer deeper than<br />
5 m at the CALM K1 site. Normally seasonal freezing<br />
penetrates from 4.5 to 5.2 m.<br />
Using data obtained from boreholes, geologic and<br />
geomorphologic data, and knowledge about the extent of<br />
snow cover and vegetation, an attempt was made to calculate<br />
spatial distribution of active-layer thickness for<br />
Zhusalykezen Mountain pass area. An area of 15x2 km*<br />
near the mountain pass was selected for the computations<br />
of depth of seasonal thaw. The deterministic model<br />
(Marchenko <strong>20</strong>01) with regular grid spacing of 50x50 m<br />
was used for computation. Figure 2 shows the calculated<br />
spatial changes in active-layer thickness within selected<br />
areas.<br />
Because in high mountain areas permafrost is one of<br />
the dominant factors affecting slope stability, knowledge<br />
about tendencies of active layer development is of great<br />
interest. Model calculations of active-layer thickness allow<br />
for the assessment of surface processes, landscape dynamics<br />
and natural hazards.<br />
REFERENCES<br />
Brown J., Hinkel K. M. and Nelson F. E. <strong>20</strong>00. The Circumpolar Active<br />
Layer Monitoring (CALM) Program: research design and initial results.<br />
Polar Geography 24 (3):165-<strong>25</strong>8.<br />
Gohunov A P. and Nemv A E. 1978. The research of temperature of<br />
loose deposits of the high mountain Tien Shan. Cryogenic Phenomena<br />
of High Mountains. Nauka, Novosibirsk: 92-99.<br />
Marchenko, S. S. <strong>20</strong>01. A model of permafrost formation and occurrences<br />
in the intracontinental mountains. Norsk Geografisk Tidsskrift<br />
- Norwegian Journal of Geography 55: 230-234.
Active layer temporal and spatial variability at European Russian<br />
Circumpolar Active Layer Monitoring (CALM) sites<br />
G. Mazhitova, "G. Ananjeva-Malkova, **O. Chesfnykh and **D. Zamolodchikov<br />
Institute of Biology, Komi Science Centre, Russian Academy of Science, Syfyvkar, Russia<br />
*Earth Cryosphere Institute, Siberian Division, Russian Academy of Sciences, Moscow, Russia<br />
""Centre for Ecology and Productivity of Forests, Russian Academy of Sciences, Moscow, Russia<br />
Three sites were established in European Russia in the<br />
framework of the "Circumpolar Active Layer Monitoring"<br />
NSF-funded project (Brown et al. <strong>20</strong>00). The Bolvansky<br />
site [R24] (68"18'N, 54"30'E) located at the western limit<br />
of the continuous permafrost zone in Europe has four<br />
years of records; Ayach-Yakha [R2] (67"35'N, 64"Il'E)<br />
with seven years of record and Talnik [R23] (67"<strong>20</strong>'N,<br />
63'44'E) with five years represent "cold" and "warm" extremes<br />
of the range of permafrost conditions in the discontinuous<br />
permafrost zone. This report was developed<br />
as a result of the CALM synthesis workshop held in November<br />
<strong>20</strong>02 at the University of Delaware.<br />
The sites represent permafrost with temperatures from<br />
-0.5 to -2.0 "C, and sensitive to even decadal-scale climatic<br />
changes (Oberman and Mazhitova <strong>20</strong>01). Data from<br />
weather stations show synchronous MAAT fluctuations in<br />
the European Russian Arctic. MAAT is -4.4"C at the<br />
maritime R24, whereas at the continental R2 and R23 it is<br />
-59°C. The difference is due to winter temperatures,<br />
whereas annual sums of positive temperatures (DDT) are<br />
equal (Fig. I).<br />
In spite of similar DDT, differences in thaw depths between<br />
the three sites occurred in all years of record; endof-season<br />
thaw averaged for the period of record was I01<br />
cm at R24 and 102 cm at R23 versus 68 cm at R2. The<br />
differences are due to effects of local factors, first of all,<br />
snow thickness. The two sites with deeper thaw, R24 and<br />
R23, reveal remarkable similarity in thaw values and dynamics<br />
(Fig. 1). Regressions of thaw depths with DDT0.5<br />
using a form of the Stefan solution (Fig 2) clearly indicate<br />
that the same increases in DDT cause much smaller increases<br />
in thaw depths at the cold R2 than at the two<br />
"warmer" sites.<br />
An inter-annual node variability (INV) index has been<br />
proposed by Hinkel and Nelson (in press) to characterize<br />
the consistency of thaw response to thermal forcing. The<br />
site average index is 10% at R23,16% at R24 and 23% at<br />
R2 ranging from 3 to 69%, from 0 to 37%, and from I to<br />
88%, respectively.The values indicate a more consistent<br />
and predictable response of the warmer R23 and R24, as<br />
compared to the cold R2. At R2, the highest INV values<br />
are characteristic for the grid nodes located in a hollow<br />
with very dynamic soil water content and in the area<br />
where bedrock is closer to the soil surface.<br />
0<br />
1300<br />
1<strong>20</strong>0<br />
1100<br />
1000<br />
900<br />
800<br />
700<br />
800<br />
Mean Amual Alr Ternperat urn, "c<br />
1995 1996 1997 lag8 1999 <strong>20</strong>00 <strong>20</strong>01 <strong>20</strong>02<br />
I I I 1 I I I I<br />
Degree Day3 Thaw,%<br />
1995 1996 1897 lssa 1999 <strong>20</strong>00 <strong>20</strong>01 <strong>20</strong>02<br />
Thaw Depth, cm<br />
Figure 1. Climate parameters and active layer depths at European,<br />
Russia's CALM sites.<br />
103
Mazhitova, E. et al.: Active layer temporal and spatial variability at European Russian. ,.<br />
Effects of various landscape factors on thaw depth Oberman, N.G. 1998. Permafrost and cryogenic processes in the<br />
were analyzed. Snow thickness, which is one of the main 540-550, East-European Russian Subarctic. Eurasian Soil Science 31(5):<br />
controls Over permafrost distribution patterns at both re- Oberman, N,G, and Mazhitova, G,G, <strong>20</strong>01, Permafrost dynamics in<br />
gional and meso-scale (Oberman 1998) may also work at the north-east of European Russia at the end of the <strong>20</strong>th century.<br />
a scale as detailed as that of 100-m CALM grids, if the Norwegian Journal of Geography 55 (4): 241-244.<br />
range of snow thickness within a site is large enough.<br />
Thus, at R24 a talik occurs in a depression under an annually<br />
developing snow drift. Surface topographic forms<br />
(ridges, basins, etc.) from several meters to several dozen<br />
meters in size and a depth to bedrock (R2) are the most<br />
powerful controls on thaw depth. The second important<br />
factor is organic layer thickness, which can be correlated<br />
to thaw depth both negatively and positively, depending<br />
on the topographic form. The positive correlation is observed<br />
at depressions and hollows where soils are saturated<br />
with water. A statistical analysis conducted for R24<br />
showed that a peat soil layer up to 22 cm thick produces a<br />
stronger effect on thaw depths than do vegetation and a<br />
sodden soil layer. Still, thaw depths under major vegetation<br />
classes differ clearly at all three sites. At R2, frost boil<br />
dynamics are correlated to average thaw depths: three<br />
cold summers in sequence with shallow thaw caused new<br />
boils to develop, whereas during the following three warm<br />
summers with deeper thaw all boils became covered with<br />
at least primitive vegetation.<br />
50<br />
30 70' "<br />
f'""";""---C-*<br />
y =0,74)<br />
914<br />
I "I-<br />
1 .I<br />
27,O 29,O 31,O 33,O 35,O 37,O<br />
Sq. root of DA, ckgrees<br />
* R F123 A E4<br />
""<br />
Linear tend (R) .- - -(F123) ..-.-( Fe4)<br />
Figure 2. End-of-season thaw depth versus square<br />
gree days.<br />
root of thaw de-<br />
Surface subsidence was assessed at R2 by repeated<br />
leveling of each grid node. Over a four-year period site<br />
subsidence averaged 3 cm in total. Thus, permafrost vertical<br />
retreat is underestimated if one bases conclusions on<br />
thaw depths alone.<br />
REFERENCES<br />
Brown, J., Hinkel, K.M. and Nelson, F.E. <strong>20</strong>00. The Circumpolar Active<br />
Layer Monitoring (CALM) Program: <strong>Research</strong> Designs and<br />
Initial Results. Polar Geography 24(3): 165-<strong>25</strong>8.<br />
Hinkel, K.M. and Nelson, F.E. in press. Spatial and Temporal Patterns<br />
of Active Layer Thickness at Circumpolar Active Layer Monitoring<br />
(CALM) sites in Northern Alaska, 1995-<strong>20</strong>00.
8th International Conference on Permafrost Extended Abstracts on Currant <strong>Research</strong> and Newly Available ~ n ~ ~ ~<br />
Comparative analysis of active layer monitoring at the CALM sites in West<br />
Siberia<br />
E.S. Melnikov, M.O. Leibman, N. G. Moskalenko and A.A. Vasiliev<br />
Earth Cryosphere Institute, Russian Academy<br />
of Sciences, Siberian Branch, Tyumen, Russia<br />
The main objective of the study is to report on recent analyses<br />
of spatial and temporal active layer variations from the<br />
CALM grids within several bio-climatic and permafrost zones<br />
of West Siberia (Melnikov et at. <strong>20</strong>01). These sites were established<br />
in the framework of the “Circumpolar Active Layer<br />
Monitoring” NSF-funded project (Brown et al. <strong>20</strong>00). In addition<br />
to the short-term CALM data, other terrain information and<br />
longer records are utilized.<br />
The zones covered by the monitoring data are typical tun<br />
dra (sites “Mar&”” and “Vaskiny Dachi”), and northern<br />
taiga (site “Nadym”). Tundra is characterized by variable<br />
landscapes of hilly plains. The northem taiga has a lesscomplicated<br />
landscape structure and generally more subdued<br />
or flatter topography compared with the typical tundra. Distinguishing<br />
features of the three CALM monitoring grids are:<br />
Mane-Sale [R3] (69043’N, 66045’E; 1000-m grid). Marine-influenced,<br />
vast areas of blowout sands, highly dissected<br />
surface, sparsely distributed peat-moss cover.<br />
Vaskiny Dachi [R5] (7001 7’N, 680 54’E; 100-m grid).<br />
Landslide affected, some blowout sands, saline clay near the<br />
surface, highly dissected surface, sparsely distributed peatmoss<br />
cover.<br />
Nadym [RI] (650<strong>20</strong>’N, 7<strong>20</strong>55’E; 100-m grid). TechnG<br />
genic impact outside the grid, permafrost occurs mainly with<br />
peatlands and frost-heave mounds, rather flat surface, thick<br />
peat and widespread moss. Zonal changes in the thickness d<br />
the active layer are controlled by a combination of factors.<br />
The summer temperature as well as the thickness of the organic<br />
mat and coverage decreases northward. The maximum<br />
active layer in peat increases respectively from 80 cm<br />
in the taiga to 50 cm in tundra, and in sand from 3 m in the<br />
taiga to 2 m in tundra. The difference in thaw rates in various<br />
climatic environments has common features with a slight rise<br />
of thaw rate northward in mineral soils associated with the re<br />
duction of thaw depth.<br />
Active layer landscape interrelation is controlled to a high<br />
degree by the composition and wetness of soils. Maximum<br />
active layer depth is characteristic of sands, especially at the<br />
blowout hollows. In addition to blowout sands, maximum active<br />
layer is found on the wettest landscapes, (1) in northem<br />
taiga, in mires and tundra with peaty hummocks, (2) in typical<br />
tundra, in ravines. Comparison of active layer in different<br />
zones is possible within similar landscape types.<br />
Marre-Sale has beenthesite of long-termobservations<br />
since 1978 (Vasiliev et al. 1998) at several grids and tmnsects.<br />
Minimum thaw depth was observed in 1999 (87 cm)<br />
followingthe low summerairtemperature. Maximum thaw<br />
depth in 1984, and 1995 (133 cm) is in agreement with high<br />
summer air temperature (6.8 to 7.0oC), while peak thaw in<br />
I990 (132 cm) was not connected with summer temperature,<br />
which was moderate (5.8oC), but was probably due to rather<br />
low summer precipitation.<br />
The Vaskiny Dachi site has been active since 1990<br />
(Leibman <strong>20</strong>01). The minimum active layer (81 cm) was ob<br />
served in 1992 (transect) and 1997 (CALM grid), and the<br />
maximum was in 1995 (98 and 94 cm, respectively). The<br />
interannual active layer variability does not directly depend<br />
on the air temperature fluctuations, but also on atmospheric<br />
precipitation and its regime, at least for the minimum active<br />
layer depth.<br />
At the Nadym transect (Moskalenko at al. <strong>20</strong>01) starting<br />
in 1972, minimum thaw was observed in 1972 and 1992 (I00<br />
cm), while maximum thaw occurred in 1989, 1995 and 1998<br />
(190cm). At the CALM grid since 1997 the maximum thaw<br />
(143 cm) was observed in a year <strong>20</strong>02, the year with rather<br />
a moderate air temperature but high summer precipitation.<br />
Minimum active layer occurred in 1999 (1<strong>25</strong> cm). There is a<br />
correlation (average coefficient -0.6) between the peat thick-<br />
ness and active layer depth at the Nadym CALM grid.<br />
Thaw depths for the tundra CALM gridsatMarre-Sale<br />
and Vaskiny Dachi are compared (Fig. 1). The trend is similar<br />
at sandy sites, though a certain difference in the active<br />
layer depths is observed. Since the climatic situation for both<br />
sites is rather similar and data is taken from the same source<br />
the explanation may be found in surface conditions reflected<br />
in ground temperature. Comparing observed ground temperature<br />
at both sites we conclude that blowing and removal<br />
of the snow in winter from the hilltops of Vaskiny Dachi de<br />
termines a rather low ground temperature, about -7 “C, while<br />
at Marre-Sale ground temperature averages -5 “C.<br />
105
Melnikov, E.S. et 81.: Comparative analysis of active layer monitoring ...<br />
160<br />
60<br />
2 4 6 a<br />
MSAT, OC<br />
I<br />
-<br />
Marre-Sale<br />
blowout<br />
depressions<br />
- IVaskiny<br />
Dachi<br />
blowout<br />
depressions<br />
- - 1<br />
Vaskiny<br />
Dachl, all<br />
sands<br />
h, cm<br />
<strong>20</strong>0<br />
150<br />
100<br />
50<br />
0<br />
Tundra<br />
Northern taiga<br />
' I "1<br />
2 4 6 8 10 12 1<br />
MSAT, "C<br />
Figure 1. Correlation between the mean summer air temperature<br />
(MSAT) and the active layer depth at the sandy surfaces of Marre-<br />
Sale and Vaskiny Dachi CALM grids.<br />
The long-term data from transects and CALM grids for<br />
tundra and taiga are presented in Figure 2. Landscape types<br />
are characterized by specific response to the atmospheric<br />
warming, independent of the bio-climatic zone. A comparison<br />
was made of mires: wetlands with water table near the surface,<br />
and peatlands with thick peat. As follows from the trend<br />
slope on Figure 2, the response of mires (the upper set d<br />
points on the diagram, Fig. 2) to possible warming is more<br />
strongly expressed as compared with peatlands (the lower<br />
set of points on the diagram, Fig. 2).<br />
Thus, temporal and spatial variations of maximum thaw<br />
depthin two bio-climatic zones of West Siberia are under<br />
study. The following conclusions are drawn.<br />
1 Various bio-climatic and permafrost zones in West Siberia<br />
show little or no warming trends during the study period.<br />
In tundra, no trend is noted. In taiga, the average trend is<br />
0.03"C per year.<br />
2 A slight trend can be discerned for thaw depth. In the<br />
typical tundra and mires of the northern taiga of West Siberia,<br />
active layer thickness increases by 0.2 % and<br />
0.5% per year, respectively. During the last several<br />
years, the north of West Siberia was characterized by<br />
drier summers; therefore the possible reason for increased<br />
depth of the active layer is atmospheric precipitation rather<br />
than air temperature changes.<br />
3 Response of the active layer to climate fluctuations differs<br />
for various landscapes. In both bio-climatic zones, peatlands<br />
are most passive due to the existence of a thick insulating<br />
organic layer. In tundra, the most active response<br />
is determined for wet terrains. In taiga, mires are the most<br />
responsive.<br />
Figure 2. Model of the climate warming: correlation between mean<br />
summer air temperature (MSAT) and active layer depth (h). Larger<br />
symbols refer to the CALM grids, smaller ones to transects; black<br />
markers refer to Marre-Sale (the tundra zone), and shaded markers<br />
are for Nadym (the taiga zone).<br />
ACKNOWLEDGEMENT<br />
This report was developed as a result of the CALM synthesis<br />
workshop held in November <strong>20</strong>02 at the University d<br />
Delaware.<br />
REFERENCES<br />
Brown, J., Hinkel, K.M. and Nelson, F.E. <strong>20</strong>00.The Circumpolar AG<br />
tive Layer Monitoring (CALM) Program: <strong>Research</strong> Designs and<br />
Initial Results. Polar Geography 24 (3): 165-<strong>25</strong>8.<br />
Leibman, M.O. Active layer dynamics and measurement technique<br />
at various landscapes of Central Yamal. Earth Cryosphere V(3):<br />
17-24 (in Russian).<br />
Melnikov, E.S., Vasiliev, AA., Malkova G.V. and Moskalenko N.G.<br />
<strong>20</strong>01. Monitoring of the active layer in the Western part of the<br />
Russian Arctic. First European Permafrost Conference, Rome,<br />
26-28th March <strong>20</strong>01. Abstracts.<br />
Moskalenko, N.G., Korostelev, Yu. V. and Chervova, E.I. <strong>20</strong>01.<br />
Monitoring of Active Layer in Northern Taiga of West Siberia.<br />
Earth Cryosphere V(1): 71-79 (in Russian).<br />
Vasiljev A.A., Korostelev Yu. V., Moskalenko N.G. and Dubrovin VA<br />
1998. Active layer monitoring in West Siberia under the CALM<br />
program (database). Earth Cryosphere<br />
sian).<br />
ll(3): 87-90 (in Rus-<br />
106
8th International Conference on permafrost Extended Abstracts an Current <strong>Research</strong> and Newly Available 1~~~~~~~~~~<br />
Recent evolution of permafrost in both Becs-de-Bosson and Lona<br />
glacierlrock glacier complexes (western Swiss Alps)<br />
S. Metrailler, R. Delaloye and *R. Lugon<br />
Dept. Geosciences, Geography, University of Fribourg, <strong>Switzerland</strong><br />
"University Institute Kurt Bosch, Sion, <strong>Switzerland</strong><br />
INTRODUCTION<br />
Geoelectrical and BTS (Bottom Temperature of the winter<br />
snow cover) measurements were repeated a decade after the<br />
first investigations on two neighbouring sites of the Valais<br />
Alps (<strong>Switzerland</strong>, 4609'N, 7031'E) where glaciers largely<br />
overrode rock glaciers during the Little Ice Age (LIA). The<br />
main objective of the present study is to provide an overview<br />
of the recent evolution of permafrost in these glacierlrock glaciercomplexes.<br />
Two different time scalesassociatedwith<br />
specific processes are involved: (a) the LIA and the impact<br />
on permafrost bodies produced by the glacier advance, and<br />
(b) the last decade of the <strong>20</strong>th century and the response d<br />
permafrost to climate warming.<br />
Tenthorey (1993) and Eerber (1994) investigated both<br />
glacierlrock glacier complexes of Becs-de-Bosson (2630 -<br />
2900 m a.s.1) and Lona (2640 - 3000 m ad.) during the period<br />
1989 to 1991, By means of geophysics, these authors<br />
observed that the permafrost distribution within the rock glaciers<br />
had been affected by the glacier advance that had occurred<br />
during the Little Ice Age (LIA).<br />
Most of the numerous vertical DC (direct current) resistivity<br />
soundings which were carried out around I990 were re<br />
peated in <strong>20</strong>02, as far as possible at the same places. The<br />
data were supplemented by further geoelectrical and BTS<br />
measurements.<br />
METHODS<br />
Twenty-eight DC resistivity soundings were performed in<br />
<strong>20</strong>02,of which ten were new ones. Contrary to the I90<br />
soundings, when only the symmetrical Schlumberger array<br />
was used, the dissymmetrical Hummel configuration was<br />
systematically applied for the more recent acquisitions. Such<br />
a configuration allowed lateral changes in the ground resistivity<br />
to be detected. It was difficult to relocate the exact location<br />
of the soundings that were camed out a decade earlier at the<br />
Becs-de-Bosson rock glacier. At Lona, Gerber (1994)<br />
marked the position of the soundings with wooden stakes.<br />
was therefore possible to reiterate exactly the same soundings.<br />
It<br />
A resistivity mapping was carried out in the rooting zone<br />
of the Becs-de-Bosson rock glacier. The apparent resistivity<br />
was investigated at a pseudodepth of about 10-15 m (<strong>20</strong> m<br />
inter-electrode spacing, Wenner configuration, 177 measurements).<br />
A BTS map based on 281 measurement points was<br />
drawn for the Becs-de-Bosson rock glacier. A similar task is<br />
planned at Lona.<br />
Finally, 30 minidata loggers (UTL-1) were dispatched an<br />
both sites in September <strong>20</strong>02 in order to control the thermal<br />
state of the ground surface during at least one year.<br />
IMPACTS OF THE LITTLE ICE AGE<br />
Becsde-Bosson<br />
The small Becs-de-Bosson glacier had completely disappeared<br />
by the middle of the <strong>20</strong>th century. During the LIA, the<br />
glacier covered the upper half of the rock glacier. Several<br />
push moraines were formed.<br />
The BTS map largely fits with geoelectrical data and<br />
therefore permits the discontinuous distribution of permafrost in<br />
the rooting zone of the rock glacier to be illustrated (Fig.1).<br />
The zone where the BTS values are above -2°C points b<br />
the current probable absence of permafrost. Most of this area<br />
was covered by the glacier during the LIA and probably at<br />
least until the end of the 19th century. Colder BTS values indicate<br />
permafrost occurrence. They correspond either to push<br />
moraines or to the intact lower part of the rock glacier. The<br />
foot of the northem steep slope of the Becs-de-Bosson peak<br />
is cold as well.<br />
h a<br />
During the LIA, the Sasseneire glacier almost completely<br />
overrode the Lona rock glacier. In some places, the glacier<br />
appears to have gone beyond the rock glacier extent. Only a<br />
few ice patches remained at the Foot of the Sasseneire headwall<br />
in <strong>20</strong>02.<br />
Geoelectric data showed that only pieces of the rock glacier<br />
have continued to exist since the LIA glacier advance.<br />
Today, there is almost no ground ice whatever left in the en-<br />
107
tire glacierlrock complex. However, permafrost was de<br />
tected, notably, in the terminal part of the rock glacier. This<br />
frozen body seems to have been pushed down by the glacier<br />
towards a more sunny area and far away from any source d<br />
debris.<br />
Thiscouldbeinterpreted as a dramaticincreaseinground<br />
temperature and probably a melting of permafrost between<br />
about 4 and 10 m depth.<br />
CONCLUSIONS<br />
The permafrost and ground ice distribution within the two glacierlrock<br />
glacier complexes reflects the glacial history over<br />
the last centuries. The investigated rock glaciers have been<br />
considerably remodelled by the LIA advance of small glaciers.<br />
Some perennially frozen materials were displaced,<br />
while others had melted completely.<br />
Over the last decade of the <strong>20</strong>th century, no tangible<br />
evolution of the permafrost geometry, that could be attributab<br />
to changing climatic conditions, was detected by means d<br />
DC resistivity soundings in both glacierlrock glacier complexes.<br />
The permafrost modifications observed at the front of the<br />
Lona glacierhock glacier complex are attributed rather to the<br />
current unfavourable location of this permafrost consecutive 03<br />
LIA glacierlrock glacier dynamic. This melting of the permafrost<br />
had begun a certain time before 1990, as Gerber (1993)<br />
already observed a settlement of about I m in the same area<br />
by comparing aerial photographs taken in 1972 and 1992.<br />
ACKNOWLEDGEMENTS<br />
Figure 1. BTS map on the upper half Becs-de-Bosson rock glacier.<br />
This project was supported by the Valaisian Society of Natural<br />
Sciences, the University Institute Kurt Bosch and the De<br />
partment of Geosciences at the University of Fribourg.<br />
CHANGES BETWEEN 1990 AND <strong>20</strong>02<br />
The comparison between the I990 and the <strong>20</strong>02 data shows<br />
an evident change for only two of the resistivity soundings.<br />
The other sixteen soundings for <strong>20</strong>02 were not significantly<br />
different than those from a decade earlier.<br />
REFERENCES<br />
Gerber E. 1994. Geomorphologie und Geomorphodynamik der<br />
Region Lona-Sasseneire (Wallis, Schweizer Alpen) unter<br />
besonderer Berucksichtigung von Lockersedimenten mit<br />
Permafrost. PhD thesis 1060. Fac. Sciences, Univ. Fribourg,<br />
<strong>Switzerland</strong> (in German).<br />
Tenthorey G. 1993. Paysage geomorphologique du Haut Val de<br />
Rechy (Valais, Suisse) et hydrologie libe aux glaciers rocheux.<br />
PhD thesis 1044. Fac. Sciences, Univ. Fribourg, <strong>Switzerland</strong> (in<br />
French).<br />
Figure 2. Vertical resistivity sounding Lo-906 repeated at a decade<br />
interval on the terminal part of the Lona glacierlrock glacier complex.<br />
The two different soundings were performed at the front d<br />
the Lona rock glacier. Similarly, they indicate a decrease d<br />
the resistivity in the uppermost former frozen layer (Fig. 2).<br />
108
Application<br />
lnsfitufe of Physical, Chemical and Biological Problems of Soil Science of the Russian Academy of Sciences, Pushchino,<br />
Moscow region, Russia<br />
INTRODUCTION its application incorrect lot of for land a MV's surfaces. are<br />
available now with the correct regional and simultaneous de<br />
When Digital Elevation Models (DEM) are used for land sur- scription of planar and profile characteristics of land surfaces<br />
face analysis, quantitative characteristics or morphometric (Shary et al. <strong>20</strong>02).<br />
variables (MVs) must be used. Physical phenomena, which<br />
could be described by MVs, might be subdivided into the Geomefrica,,andfoms<br />
following 4 groups:<br />
Morphometric conditions of surface run& and soil through- Geometrical - landforms are land surface forms which do not<br />
flow. need to take gravitational<br />
Geometrical landforms. tion. For example, ridge and valley forms are referred to as<br />
Morphometric conditions of thermal regimes on slopes. geometrical forms if they are defined independently from their<br />
Morphometric conditions of altitude zonality, i.e.,the change orientation in gravitational fields. Locally, maximum and<br />
of atmospheric properties with elevation. minimum curvatures describe such forms.<br />
All 4 groups occur in the cryolithic zone.<br />
MORPHOMETRIC VARIABLES<br />
The land surface can be described by regional and local<br />
MVs. In the first case, the influence of remote sites is taken<br />
into account, in the second case it is not. In this study, a<br />
system of 18 MVs of general geomorphometry was used<br />
(Shary et al. <strong>20</strong>02).<br />
Morphometric conditions of runoff<br />
The planar land surface is regionally described by maximum<br />
catchment area (MCA) and maximum dispersal area (MDA).<br />
MCA for a given matrix element (square on a map) shows<br />
the maximum area from which material moving downslope<br />
may be collected. MDA describes the maximum area on<br />
which materials moving downslope may be diffused.<br />
The planar dimension and profile of the land surface are<br />
locally described by horizontal (kh) and vertical (kv) curvatures.<br />
The differentiation of land surfaces with respect to kv<br />
and kh allows us to group them into four local landform types<br />
(Troeh 1964).<br />
Algorithms are frequently used which integrate slope gradient<br />
and aspect (GA) to simultaneously describe both planar<br />
and profile characteristics. The most popular formula is the b<br />
pographical index (Beven and Kirkby 1979). This method,<br />
however, tends to approach infinity with GA-0 and makes<br />
APPLICATION OF MORPHOMETRIC VARIABLES<br />
Here we consider the application areas of an MV system in<br />
geocryology but have not considered the morphometric conditions<br />
of altitude zonality and of the thermal regimes d<br />
slopes. The primary possibility concerns the analysis of tunoff<br />
directions and potential flowspeeds (movement of water,<br />
avalanches, glaciers, lava, etc.).<br />
The MCA map reflects both potential and real hydrological<br />
channel networks without distinguishing them (Fig. 1 and 2);<br />
during calculation, the locations and depths of all depressions<br />
were also computed. This MV is of a critical value and de<br />
termines free water at the surface. The places of output of<br />
surface waters and their areas of their origin are well described.<br />
For the definition of potential runoff rates it is necessary b<br />
take the slope profile along the length of the flow path into<br />
consideration. Areas of relative flow deceleration always coincide<br />
with the concave parts of the profiles, while areas d<br />
relative flow acceleration correspond to the convex parts.<br />
The development of cryogenic processes such as thermokarst<br />
and solifluction, the formation of frost mounds and<br />
alas etc. is connected with theoccurrenceofwater in the<br />
soillground. This allows the prediction of position and potential<br />
intensity based on the correlation of the soil/ground moisture<br />
content (and some other parameters) with the MVs. It is important<br />
to note that in some of the terrains from the 18 studied<br />
109
MVs, only the geometrical form provided a strong correlation<br />
with soil moisture.<br />
Already existing cryogenic landforms can be found as a<br />
result of morphometric analysis of DEMs. In general, landforms<br />
of different genesis can be found. Landforms with gee<br />
morphometric structures typical for cryogenic landforms can<br />
be separated from other landforms but only by the application<br />
of all 18 MVs.<br />
Mitusov, A. V and Mitusava, 0. E.: Application of quantitative land-surface.. .<br />
forecast the potential imgation a of territory and to estimate he<br />
hazards.<br />
(2) predict the places of development of cryogenic pro^<br />
esses (thermokarst, solifluction, formation of frost mounds and<br />
alases, etc.) as well as their potential intensity.<br />
(3) reveal cryogenic landforms on DEM and separate<br />
them from other landforms (geological, etc.).<br />
ACKNOWLEDGEMENTS<br />
The authors are grateful to P.A. Shary for providing Analytical<br />
GIS Eco with data and Dr 0.1. Khudyakov for advice in<br />
the geocryology field.<br />
REFERENCES<br />
Beven, K.J. and Kirkby, M.J. 1979. A physically based, variable contributing<br />
area model of basin hydrology. Hydrological Sciences<br />
Bulletin 24: 43-69.<br />
Shary, P.A., Sharaya, L.S., and Mitusov, A.V. <strong>20</strong>02. Fundamental<br />
quantitative methods of land surface analysis. Geoderma 107<br />
(1-2): 1-35.<br />
Troeh, F.R. 1964. Landform parameters correlated to soil drainage.<br />
Soil Science Society of America <strong>Proceedings</strong> 28: 808-812.<br />
1062 1 ho<br />
338 s<br />
106 s<br />
33 9<br />
101<br />
3 dl<br />
1 0%<br />
U 343<br />
0 109<br />
003&1<br />
onlo9<br />
0 0031<br />
OBO!!<br />
0 00034<br />
00OOlU<br />
0 OOOOI<br />
0<br />
unlmownv<br />
Figure 2. Example of a matrix with MCA calculated from the DEM<br />
given in Figure 1. The map of MCA allows to predict the location of<br />
rivers. It follows from comparison between Figure 1 and Figure 2<br />
that the locations of rivers coincides with areas of the large MCAs.<br />
CONCLUSIONS<br />
The application of an extended system ofquantitativeland<br />
surface analysis methods in cryology - except for the morphometric<br />
conditions of thermal regimes on slopes and altitude<br />
zonality - allows us to:<br />
(1) define a direction and potential speed of flow movements<br />
for various substances (water, glaciers, lava, landslides,<br />
snow avalanches, etc.); on this basis it is possible k~<br />
110
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available ~ ~ f<br />
Modeling permafrost distribution in the northern Alps using global<br />
radiation<br />
0. Mustafa, M. Gude and *M. Hoelzle<br />
Department of Geography, University of Jena, Germany<br />
"Glaciology and Geomorphodynamics Group, Department of Geography University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
INTRODUCTION<br />
In order to reduce natural hazards in alpine areas high<br />
mountain permafrost became an important objective of research<br />
within the last decade. Many investigations were<br />
made in the central Alps. Here, the spatial distribution of<br />
permafrost is well documented because a high number of<br />
field measurements and computer models give reasonable<br />
results. But the adaptation of these findings to the<br />
northern Alps margin is assumed to cause problems due<br />
to the different climatic conditions controlling the permafrost<br />
regime. A regional analysis should especially account<br />
for the higher diffuse radiation compared to direct radiation.<br />
Therefore, the CLOUDMAP model is developed as a<br />
variant of the PERMAMAP model.<br />
THE TEST SITES<br />
The investigation area in the mountains of Wetterstein is<br />
situated at the northern margin of the Alps at 47"<strong>25</strong>'N and<br />
IO"59'E. It is crowned by Germany's highest peak, the<br />
Zugspitze (2962 m a.s.1.) and the lowest elevation is at<br />
1580 m a.s.1. A <strong>20</strong> m grid is used for model runs at this<br />
site. For calibration and comparison a secondary test site<br />
in the central Alps south of Zermatt (Valais) was chosen<br />
(47"02'N, 7'45'E). The <strong>25</strong> m DEM of this site reaches<br />
from 1675 to 3424 m a.s.1. Climatic parameters with relevance<br />
to the models are listed in Table I.<br />
Table 1 : Climatic parameters of the test sites.<br />
Wetterstein Zermatt<br />
0°C-lsotherme [m NN] 2101 2363<br />
Temp. gradient [WIOO m] 0.60 0.51<br />
Daily sunshine [h] 5.7 6.0<br />
Cloudiness [I1101 6.9 4.1<br />
The Zermatt test site has, for years, been the site of<br />
intensive permafrost research. A great deal of field data<br />
are collected and various computer models were applied<br />
(e.g. Gruber and Hoelzle <strong>20</strong>01). In contrast, there is only<br />
little information about the permafrost situation at the<br />
Wetterstein test site (Ulrich and King 1993).<br />
Permafrost-mapping was accomplished by long-term<br />
measurements of BTS and bedrock temperature and<br />
analysis of observed permafrost and permafrost related<br />
features (Gude and Barsch <strong>20</strong>03). For spatial extrapolation<br />
of the measurements, GIS-based modeling was undertaken<br />
with two different models, which will be presented.<br />
PERMAFROST DISTRIBUTION MODELS<br />
The PERMAMAP model (Hoelzle 1994) calculates the<br />
spatial distribution of alpine permafrost by applying a statistical<br />
relation of field measurements of BTS (bottom<br />
temperature of snow) with elevation and direct solar radiation<br />
during summer.<br />
To account for the more humid and, therefore, more<br />
cloudy conditions in the northern Alps the CLOUDMAP<br />
model was developed. CLOUDMAP relates the BTS to<br />
elevation and global radiation (Fig. 1). An approach by<br />
Arck & Escher-Vetter (1997) is the basis for the calculation<br />
of global radiation. In contrast to the PERMAMAP model,<br />
CLOUDMAP needs the input of meteorological data for<br />
cloudiness and duration of sunshine.<br />
RESULTS<br />
Both models were applied to both test sites. Due to the inaccessibility<br />
of large parts of the Wetterstein during winter<br />
it was not possible to get enough BTS-data to generate<br />
the statistical relations the models need. Hence, a regional<br />
calibration by climatic parameters was applied to the statistical<br />
relations of the Zermatt test site in order to transfer<br />
them to Wetterstein.<br />
A comparison of the model results for the Zermatt test<br />
site shows only minimal differences. In contrast, the results<br />
for the Zugspitze test site differ significantly (Fig. 2).<br />
The difference concerns most notably the north-exposed<br />
slopes, where the lower limit of permafrost rises about <strong>25</strong>0<br />
m of elevation.<br />
111
Mustafa, 0. et al.: Modeling permafrost distribution in the northern Alps<br />
........................<br />
..........................<br />
MAAT<br />
Global Radtatlon<br />
Ij<br />
.........................<br />
Input j<br />
...<br />
I<br />
Output<br />
Figure 1. Structure of CLOUDMAP as an insertion to PERMAMAP; fl-3(X) - empirical functions<br />
Figure 2. Modelled permafrost distribution at the Wetterstein test site (glaciers are not considered).<br />
REFERENCES<br />
Arck, M. and Escher-Vetter, H. 1997. Topoclimatological analysis of<br />
the reduction of the glaciers in the Zugspitz region, Bavaria. Zeitschrift<br />
fur Gletscherkunde und Glazialgeologie lX(1-2): 57 72.<br />
Gruber, S. and Hoelzle, M. <strong>20</strong>01. Statistical modelling of mountain<br />
permafrost distribution: Local calibration and incorporation of remotely<br />
sensed data. Permafrost and Periglacial Processes 12(1):<br />
69 - 77.<br />
Gude, M. and Barsch, D. <strong>20</strong>03 (in print). Assessment of geomorphic<br />
hazards in connection with permafrost occurrence in the Zugspitze<br />
area (Bavarian Alps). Geomorphology,<br />
Hoelzle, M. 1994. Permafrost und Gletscher im Oberengadin. Grundlagen<br />
und Anwendungsbeispiele fur automatisierte Schatzverfahren.<br />
Mitteilungen der VAW 132. ETH <strong>Zurich</strong>.<br />
Ulrich, R. and King, L. 1993. Influence of mountain permafrost on<br />
construction in the Zugspitze mountains, Bavarian Alps, Germany.<br />
Sixth International Conference on Permafrost, Beijing, <strong>Proceedings</strong><br />
1 : 6<strong>25</strong> - 630.<br />
112
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available ~ ~ ~<br />
The circumpolar active layer monitoring (CALM) program: Recent<br />
Developments<br />
F. E. Nelson, N. 1. Shiklomanov, *K. M. Hinkel and **J. Brown<br />
Department of Geography, University of Delaware, Newark, DE USA<br />
“Department of Geography, University of Cincinnati, Cincinnati, OH USA<br />
**International Permafrost Association, Woods Hole, MA USA<br />
The Circumpolar Active Layer Monitoring (CALM) program (<strong>20</strong>00). Under the aegis of the IPA’s Permafrost and Global<br />
was established to observe the long-term response of the aG Change Working Group, meetings and discussions devoted<br />
tive layer and near-surface permafrost to changes in climatic to CALM have been held at most of the annual cryosphere<br />
parameters. Initial foci of the program included “data rescue” conferences in Pushchino, Russia and at several of the Ameactivities<br />
and creation of a data archive, building an alliance d rican Geophysical Union’s Fall Meetings in San Francisco.<br />
active field scientists willing to share data, execution of critical<br />
The current phase of CALM, supported by the U.S. National<br />
field experiments, and development of data-collection prote<br />
Science Foundation (NSF), reached its apogee in November<br />
cols. Most aspects of these tasks were implemented by the<br />
mid-1990s. Details are provided in the volume by Brown et<br />
<strong>20</strong>02, when 35 scientists from six countries attended an<br />
al. (<strong>20</strong>00).<br />
NSF-sponsored workshop in Lewes, Delaware. This<br />
Data produced from the CALM network have proven workshop provided an opportunity for CALM scientists to<br />
useful in many contexts, including model validation. Owing 13 learn details about research programs at various sites, to dithe<br />
limited observational record at most sites, it is not yet scuss progress and problems in the network, to implement<br />
possible to arrive at definitive conclusions about long-term unified data-analytic procedures, and to plan future activities.<br />
changes or trends in the temperature and thickness of the ac- The CALM network currently includes 130 active sites<br />
tive layer. The few long-term data sets available from high- with 15 participating countries. An additional <strong>20</strong> secondary<br />
latitude sites in the Northern Hemisphere show very sub- sites are cc-located with these primary sites, resulting in a<br />
stantial interannual and interdecadal fluctuations. The spatial total of 150 sites that report active-layer thickness on an ak<br />
variability of active-layer thickness is large, even within gm<br />
nual basis. Of the total, 100 sites are collecting soil and pergraphicareas<br />
of very limited extent (e.g. 1 km2). Most sites<br />
mafrost temperature data; of these, 60 are from boreholes<br />
in tundra environments show a strong, positive relation be<br />
tween summer temperature and the thickness of the active<br />
layer. This relation is weaker in boreal environments. Increases<br />
in thaw penetration, subsidence, and development of<br />
thermokarst terrain have been observed at some sites, particularly<br />
in subarctic regions.<br />
Strong bonds have been forged between CALM and several<br />
international monitoring and global change programs.<br />
CALM is part of the International Permafrost Association’s<br />
Global Terrestrial Network for Permafrost (GTN-P), which<br />
between 3 m and 884 m deep (Smith et al, this volume,<br />
discuss metadata information on GTN-P boreholes). Most<br />
sites in the CALM network are located in Arctic and Subarctic<br />
lowlands, although <strong>20</strong> boreholes are in mountainous regions<br />
of the Northern Hemisphere above 1300 m elevation. A<br />
new Antarctic component is forming and currently contains 13<br />
sites.<br />
The <strong>20</strong>02 CALM workshop focused synthesis activities<br />
on the 70 sites in six countries from which gridded data are<br />
also incorporates a network of boreholes for monitoring per- available. The workshop resulted in preparation of seven exmafrost<br />
temperatures (Romanovsky et al. <strong>20</strong>02). GTN-P is tended regional abstracts for this volume (Table I). Several<br />
itself part of the Global Terrestrial Observing System (GTOS) other papers report details from these same CALM sites<br />
and the Global Climate Observing System (GCOS), co- (Romanovsky et ai. <strong>20</strong>02, Hinkel and Nelson <strong>20</strong>03) Other<br />
sponsored by a consortium of international organizations, CALM related papers appear in the <strong>Proceedings</strong> and<br />
including the World Meteorological Organization and the Inter- Abstract volumes for Mongolia (Sharkhuu), Kazakhstan<br />
national Council of Scientific Unions, among others. Further (Marchenko), Canada (Tarnocai et ai.), and Antarctic<br />
details are provided in Burgess et al. (<strong>20</strong>00).<br />
(Guglielmim et al.).<br />
Only a little more than a decade into its existence, CALM Extended discussions about CALM’S future directions,<br />
has produced a large body of scientific literature, much of it strategies, and activities took place at the Delaware<br />
reviewed in a monograph-length paper by Brown et al. workshop, culminating in the CALM Workshop Resolution.<br />
113
Nelson F.E. et 4.: The circumpolar active layer monitoring (CALM) program ...<br />
The Resolution consists of the following main recommendations:<br />
To continue the CALM observations and improve the<br />
thematic representativeness of the network;<br />
To improve linkages with other networks and international<br />
programmes such as WCRP (CIiC), IGBP and the proposed<br />
Circumarctic Environmental Observatories Network<br />
(CEON);<br />
To better understand active layer dynamics through ob<br />
servations of borehole temperatures, subsidence, snow<br />
cover, soil properties (moisture, surface organics, texture),<br />
topography and vegetation;<br />
To ensure continued operation, maintenance and en-<br />
hancement of CALM and borehole temperature databases<br />
(GTN-P) and websites;<br />
To provide input for improvement of permafrostdimate<br />
model development, model result comparison, and verification;<br />
To prepare reports to disseminate results of data analy-<br />
ses beginning with the 8th ICOP and a special journal<br />
publication; and<br />
To develop strategies for remote sensing-based methods<br />
of monitoring geocryological parameters over extensive<br />
areas, consistent with the GHOST observation scheme<br />
Table 1. Extended abstracts resulting from <strong>20</strong>02 CALM workshop<br />
(senior author listed for abstract in this volume).<br />
1. Nordic Region: Christiansen et al.<br />
2. Lower Kolyma River: Fyodorov-Davydov et al.<br />
3. West Siberia: Melnikov et al.<br />
4. European Russia: Mazhitova et al.<br />
5. West Siberia: Vasiliev et al.<br />
6. Chukotka: Zamolodchikov et al.<br />
7. Modelling: Oelke and Zhang<br />
Romanovsky, V., Smith, S., Yoshikawa, K. and Brown, J. <strong>20</strong>02. Permafrost<br />
temperature records: indicators of climate change. Eos,<br />
Transactions, American Geophysical Union 83(50): 589 and<br />
593-594.<br />
ACKNOWLEDGEMENTS<br />
The CALM program work was supported by the U.S. National<br />
Science Foundation under grants OPP-973<strong>20</strong>51,<br />
0094769, 0095088, and 02<strong>25</strong>603. Views expressed in this<br />
abstract are the authors' and do not necessarily reflect those<br />
of NSF.<br />
REFERENCES<br />
Brown, J., Hinkel, K.M. and Nelson, F.E. <strong>20</strong>00. The Circumpolar Active<br />
Layer Monitoring (CALM) program: research designs and<br />
initial results. Polar Geography 24(3): 165-<strong>25</strong>8.<br />
Burgess, M., Smith, S.L., Brown, J., Romanovsky, V. and Hinkel, K<br />
<strong>20</strong>00. Global Terrestrial Network for Permafrost (GTNet-P):<br />
permafrost monitoring contributing to global climate observations.<br />
Geological Survey of Canada, Current <strong>Research</strong> <strong>20</strong>00-<br />
E14.<br />
Hinkel, K.M. and Nelson, F.E. in press. Spatial and temporal patterns<br />
of active layer thickness at circumpolar active layer monitoring<br />
(CALM) sites in northern Alaska, 1995-<strong>20</strong>00. Journal of<br />
Geophysical <strong>Research</strong>-Atmospheres.
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
K.A.Novotofshaya-Vlasova, D.A.Gi/ichinsky, *A.A. Abramov and **V.S. Soina<br />
Soil Cryology Laboratory, Institute of Physicochemical and Biological Problems in Soil Science, Pushchino, Moscow Region,<br />
Russia<br />
“Department of Geocryology, Moscow State University, Moscow, Russia<br />
** Department of Soil Biology, Moscow State University, Moscow, Russia<br />
Sterile frozen samples were taken from boreholes (Fig. 1)<br />
near the Laptev Sea coast (East Siberia) and in Beacon<br />
Valley (Antarctica) for iiclassical” microbiological analyses.<br />
The examined Arctic sediments were formed about 10400<br />
thousand years ago (Schirrmeister et al. <strong>20</strong>02). The sediments<br />
are ice-rich (up to 80-90%) and have a low mean annual<br />
temperature (-10 to -15T). We also investigated ancient<br />
Antarctic permafrost, several million years old (Sugden et al.<br />
1995) taken from boreholes drilled in the region of the Dry<br />
Valleys, which have mean annual temperatures ranging from<br />
-<strong>20</strong> to -22°C with ice contents lower than in Arctic deposits<br />
(40%). The magnitude of unfrozen water and total organic<br />
matter content (TOC) in Arctic loam is 33%. In Antarctic<br />
sands, both these values are close to zero, i.e., at levels impossible<br />
to record with presently available instrumental methods.<br />
Special procedures were used to determine the biological<br />
activity in the thawed samples. The number of viable cells<br />
was counted using the plate method. In the model experiments,<br />
the long-term viability of isolated species of both<br />
spore-forming and non spore-forming bacteria was maintained,<br />
for I to 4 months in a diluted nutrient medium, that is<br />
tryptic soy broth and water suspensions at room and lower<br />
(7°C) temperatures.<br />
The communities of micro-organisms inhabiting the ancient<br />
permafrost could be easily recovered. These communities<br />
are chiefly represented by bacteria concentrations varying<br />
from 102 to 107 cellslgdw (Fig. 1).<br />
Taxonomic identification of isolated aerobic bacteria<br />
showed that such biotopes contain the same diversity of<br />
bacteria as modern tundra cryosols. The taxonomic diversity<br />
revealed the tendency to survive predominantly as non<br />
spore-forming, presumably psychrotrophic bacteria able b<br />
grow on nutrient media at temperatures of 4 to 22°C. Increase<br />
in the microbial (=permafrost) age or decrease in temperature<br />
of Antarctic sediments in comparison with Arctic<br />
ones result in the decrease of proliferating cell numbers. This<br />
suggests that part of the permafrost microbial community may<br />
either undergo deep anabiosis, or be preserved in stable, socalled<br />
viable, but non-cultivable forms. As mentionedin an<br />
earlier study (Vorobyova et al. 1997), the ratio between cultivable<br />
and non-cultivable microbial cells might indicate the<br />
physiological response of microorganisms to environmental<br />
conditions.<br />
The ability of bacteria to survive in frozen sediments may<br />
be explained by specific mechanisms of the cell differentiation<br />
into forms that enable bacteria to sustain conditions unfavorable<br />
for growth and multiplication (cold stress, starvation,<br />
etc.). The colony-forming units (CFU) in long-stored cultures<br />
decreased from 108 down to 102. Nevertheless, as may be<br />
seen by electron microscopy, the majority of the cells are intact<br />
and resume their activity without additional procedures in<br />
conditions favorable for growth. Bacteria Bacillus sp., and<br />
Azotobacfersp., isolated from permafrost, are known to pre<br />
duce typical resting forms - spores and cysts in their life cycles.<br />
However, for long-term storage under conditions of<br />
starvation and subzero temperatures, they produce other<br />
forms (Fig. 2). Such specific cell forms as observed in the<br />
stored cultures may be regarded as one of the reversible<br />
dormant types of both, non spore-forming bacteria and bacteria<br />
which are known to produce specific resting cells that can<br />
beformed in permafrost. Probably, this mechanism of survival<br />
in extreme conditions might also work in an extraterrestrial<br />
environment. The investigated deposits may be considered<br />
approximately similar to Martian permafrost which,<br />
according to the most recent data, contains ground water ice<br />
(Mitrofanov et al. <strong>20</strong>02). The existence of thermo-resistant viable<br />
bacterial cells with specific ultrastructure (similar to dormant<br />
forms) in terrestrial permafrost is a reason for advancing<br />
the hypothesis that Martian permafrost might contain Life.<br />
115<br />
ACKNOWLEDGEMENT<br />
This work was supported by the Russian Foundation for Basic<br />
<strong>Research</strong>, grant 01-05-65043.
REFERENCES<br />
Gilichinsky D. <strong>20</strong>02. Permafrost as a microbial habitat. Encyclopedia<br />
of Environmental Microbiology,Willey: 932-956.<br />
Mitrofanov I. et al. <strong>20</strong>02. Maps of Subsurface Hydrogen from the<br />
High Energy Neutron Detector, Mars Odyssey. Science 297: 78-<br />
81.<br />
Schirrmeister L. et al. <strong>20</strong>02. Paleoenviromental and paleoclimatic<br />
records from permafrost deposits in the Arctic region of Northern<br />
Siberia. Quaternary International 89: 97-118.<br />
Sugden D. et al 1995. Preservation of Miocene glacier ice in East<br />
Antarctica. Nature 376: 412-414.<br />
Vorobyova E. et al. 1997. The deep cold biosphere: facts and hypothesis.<br />
FEMS Microbiology Reviews <strong>20</strong>: 277-290.<br />
116
8th International Conference an Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Comparing thaw depths measured at CALM field sites with estimates from<br />
a medium-resolution hemispheric heat conduction model<br />
C. Oelke and T. Zhang<br />
National Snow and Ice Data Center, Crysospheric and Polar Processes Division, ClR€S, University Of Colorado,<br />
Boulder, Colorado, U.S.A.<br />
INTRODUCTION<br />
The Circumpolar Active Layer Monitoring (CALM) program<br />
(Brown et al. <strong>20</strong>00) was initiated in the late 1980s<br />
to observe and understand the response of the active<br />
layer and near-surface permafrost to climate change.<br />
This systematic study focuses on active layer processes<br />
and monitoring thaw depth. It currently incorporates<br />
measurements from more than 100 sites involving 15 investigating<br />
countries. The CALM site data were collected<br />
primarily over 100 m x 100 m or I000 m x 1000 m<br />
test areas. CALM data are available from the CALM<br />
website at htf~://~2.~issa.uc.edu/-kenhinke/CALMI.<br />
Here we compare these field measurements of thaw<br />
depth with results from a numerical heat conduction<br />
model.<br />
Heat Conduction Model<br />
A finite difference model for one-dimensional heat conduction<br />
with phase change (Goodrich 1982) is used.<br />
This model has been shown to provide excellent results<br />
for active layer depth over permafrost and soil temperatures<br />
(Zhang et al. 1996) when driven with well-known<br />
boundary conditions and forcing parameters at specific<br />
locations. A detailed description of the model is given by<br />
Goodrich (1982) and Zhang et al. (1996). Our intent is to<br />
model the soil freezelthaw state for the whole Arctic<br />
drainage area. We run the model one-dimensionally and<br />
assume no lateral heat transfer among the <strong>25</strong> km x <strong>25</strong><br />
km grid cells with depths of 15 m. Soil is divided into<br />
three major layers (0-30 cm, 30-80 cm, and 80-1500 cm)<br />
with distinct thermal properties of frozen and thawed<br />
soil, respectively. Calculations are performed on 54<br />
model nodes ranging from a thickness of 10 cm near the<br />
surface to I m at 15 m depth. Details about model settings<br />
and the forcing data sets (surface air temperature,<br />
snow thickness, soil bulk density, soil moisture content)<br />
are given by Oelke et al. (<strong>20</strong>03). The model is spun up<br />
for 50 years in order to obtain more realistic start conditions<br />
for temperatures of all model layers. The model is<br />
applied to the whole Arctic area drainage comprising<br />
39,926 north polar Equal-Area Scalable Earth (EASE)<br />
grid cells with an area of 24.7~106 km2. We use a daily<br />
time step for the 22 years of our simulation (1980<br />
through <strong>20</strong>01). Oelke et al. (<strong>20</strong>03) show fields of thaw<br />
depth for this area at three-month intervals, together with<br />
the active layer depth and the length of freezing and<br />
thawing periods.<br />
COMPARISON WITH CALM THAW DEPTH<br />
MEASUREMENTS<br />
Arctic Drainage Area<br />
We compare our model results for the year <strong>20</strong>00 with<br />
measurements of thaw depth (DT) for the same year,<br />
made at 66 locations in regions of continuous or discontinuous<br />
permafrost. Measurements from Canada,<br />
Greenland, Russia, Svalbard, and Alaska are included.<br />
Figure I shows the comparison for the day when the<br />
measurement was taken in the field, which is usually<br />
close to the annual late summer maximum. Also shown<br />
are bars of +_I standard deviation for 48 grid sites. Modeled<br />
thaw depth is a distance-weighed average between<br />
the four grid cells around the CALM site location.<br />
The comparison between modeled thaw depths<br />
based on 6<strong>25</strong> km2 averages of temperature, snow<br />
depth, and soil bulk density and a measured thaw depth<br />
on a much smaller scale (1 ha or I km2) should be<br />
viewed as an indication of model performance rather<br />
than a true validation. The modeled values nevertheless<br />
agree reasonably well with the measured values within<br />
their standard deviation, and capture the general picture<br />
of high-Arctic active layers. The high standard deviation<br />
of the grid measurements points to large spatial variability<br />
of thaw depth, even on these small scales.<br />
Grid cells with modeled thaw depths much lower than<br />
measured depths are located in Greenland, Svalbard, or<br />
Baffin Island fjords at lower elevations than those of the<br />
higher, and consequently colder, corresponding model<br />
grid cells. The resulting modeled thawing seasons are<br />
shorter and the thaw depths are shallower. The RMS er-<br />
117
or without seven such sites (marked by diamonds in<br />
Figure 1) is reduced from 32.1 cm to 23.7cm. The comparison<br />
to 70 CALM sites for 1999 shows a similar behavior<br />
as for <strong>20</strong>00, with a RMS error of 29.7 cm. lt is reduced<br />
to 26.7 cm without six sites that experience a<br />
strong cold bias adjacent to mountain ranges.<br />
Qelke, C. et ai.: Comparing thaw depths measured at CALM filed sites ...<br />
the analysis in 1995 with rul=0.961 and ru2=0.936. Differences<br />
in U1 and U2 thaw depths are up to 2 cm, due<br />
to the different location and size of the sites, and different<br />
probing dates.<br />
Measured and modeled thaw depths agree quite<br />
well, taking into account the considerable differences in<br />
scales and the large spatial variability measured on the<br />
very much smaller CALM grids.<br />
madeled D,<br />
0.0 0.5 1.0 1.5 2.0<br />
Modeled DT (m)<br />
Figure 1. Modeled vs. measured thaw depth (DT) for the day when<br />
the measured value was observed in <strong>20</strong>00. Bars give the *I standard<br />
deviation for CALM sites measured on grids. Diamonds mark<br />
sites adjacent to mountains or high plateaus; therefore, modeled<br />
thaw depths are too shallow.<br />
Local Time Series<br />
The comparison between modeled and measured<br />
thaw depth for the two CALM sites near Barrow (north<br />
slope of Alaska at 71' 19'N and 156' 36'W) shows DT<br />
close to the diagonal at about 0.3 m (triangles in Figure<br />
I). A 22-year time series (1980-<strong>20</strong>01) of active layer<br />
depth (ALD) for the model grid cells near Barrow is<br />
shown in Figure 2. Only a very small positive linear trend<br />
of 0.029 cmla is calculated. The solid bold line gives<br />
thaw depth (DT) for the day of year the CALM measurements<br />
were taken, and is therefore lower than the ALD.<br />
This modeled thaw depth is compared to measurements<br />
at CALM sites U1 (Barrow, I000 m grid, dotted line) and<br />
U2 (Barrow, CRREL 10 m plots, dashed line). Both<br />
modeled and measured values show an increase in<br />
thaw depth from 1991 to 1994. The model value then<br />
decreases before rising to a maximum in 1998. From<br />
1998 to <strong>20</strong>01 there is a steady decline in thaw depth at<br />
Barrow. A correlation coefficient rul of 0.421 is calculated<br />
between modeled and measured thaw depth at U1<br />
(1992-<strong>20</strong>01). For the site U2, Ru2 is slightly higher at<br />
0.609 (1991-<strong>20</strong>01). Modeled and measured curves are<br />
almost parallel after 1994, with differences up to 5 cm.<br />
The correlations become much stronger when starting<br />
0.0<br />
1980 1985 1990 1995 <strong>20</strong>00<br />
Yea<br />
Figure 2. Modeled active layer depth (ALD) at Barrow, Alaska, for<br />
1980-<strong>20</strong>01, including the linear trend (dash-dotted line). Bold dotted<br />
lines are measured thaw depths (DT) at CALM sites U1<br />
(1992-<strong>20</strong>01) and U2 (1991-<strong>20</strong>01), and the solid bold line is the<br />
modeled thaw depth for the actual day of measurement.<br />
ACKNOWLEDGMENTS<br />
This report was developed as a result of the CALM synthesis<br />
workshop held in November <strong>20</strong>02 at the University<br />
of Delaware. This study was supported by NASA<br />
through grant NAG568<strong>20</strong> and by NSF through grant<br />
OPP-9907541.<br />
REFERENCES<br />
Brown, J., Hinkel, K.M. and Nelson, F.E. <strong>20</strong>00. The Circumpolar<br />
Active Layer Monitoring (CALM) Program: <strong>Research</strong> Designs<br />
and Initial Results, Polar Geography, 24(3): 163-<strong>25</strong>8.<br />
Goodrich, L.E. 1982. Efficient numerical technique for onedimensional<br />
thermal problems with phase change, International<br />
Journal of Heat and Mass Transfer, 21: 615421,<br />
Oelke, C., Zhang, T. Serreze, M. and Armstrong. R. <strong>20</strong>03. Regional-scale<br />
modeling of soil freezelthaw over the Arctic drainage<br />
basin, Journal of Geophysical <strong>Research</strong> 108(D10): 4314,<br />
doi: 10.10291<strong>20</strong>02JD002722..<br />
Zhang, T., Osterkamp, T.E. and Stamnes, K. 1996. Influence of the<br />
depth hoar layer of the seasonal snow cover on the ground<br />
thermal regime, Water Resources <strong>Research</strong>, 32(7):<br />
<strong>20</strong>75-<strong>20</strong>86.<br />
118
8th International Conference on Permafrost Extended Abstracts an Current <strong>Research</strong> and Newly Available I ~ ~<br />
Coastal dynamics of the Pechora Sea under human impact<br />
SA. Ogorodov<br />
Faculty of Geography, Moscow State University, Moscow, Russia<br />
The so-called Pechora Sea is located in the south-eastern<br />
part of the Barents Sea. The Pechora sector of the<br />
Barents Sea is characterised by loose permafrost coasts.<br />
These coasts are relatively stable under natural conditions<br />
but are rapidly destroyed under human impact. A case in<br />
point is the Varandei industrial area where expeditious<br />
measures to protect industrial buildings and living<br />
accomodation are necessary. Human impact upon the<br />
Varandei area activates coastal erosion because of<br />
improper exploitation that does not take peculiarities of<br />
coastal relief and dynamics into account. Coastal erosion<br />
of the Varandei area brings a threat to the settlement, oil<br />
terminal (Fig. I) and airport.<br />
Active exploitation of the Varandei industrial area and<br />
Varandei Island started in the seventies. Varandei Island<br />
is a typical sand barrier beach in the Pechora sea. The<br />
shoreface of this accumulative form is covered with a<br />
dune belt (avandune) of 4 to 7 m high. Laidas or highwater<br />
surge berms occupy the inner part of the barrier<br />
beach behind the dune belt. Varandei Island was subjectted<br />
to very strong technogenic impacts. The main industrial<br />
base was built and Novyi Varandei, a settlement for<br />
3.5 thousand inhabitants. A well-drained dune belt of the<br />
barrier beach, made up of sand beds with low ice content,<br />
was chosen as a place for the settlement, oil terminal and<br />
storehouses because it seemed to be more stable from<br />
the engineering-geological point of view than the<br />
surrounding swampy tundra lowland.<br />
Construction of the settlement and industrial base,<br />
along the edge of the erosional coastal cliff, was<br />
acompanied by repeated removal of sand and sandpebble<br />
sediments for building materials from the<br />
avandune and beach. This is absolutely wrong for the<br />
zones of wave energy divergence, especially in those that<br />
already suffer erosion.<br />
Within the zone of industrial exploitation, the coastal<br />
bluff and the coastal zone experienced considerable<br />
mechanical deformations of landforms because of<br />
transport, mechanical leveling of coastal declivities and<br />
other technogenic disturbances. A systemless use of<br />
transportation and construction techniques, including<br />
caterpillars, caused degradation of soil and plant cover of<br />
the whole dune belt of Varandei Island. Under conditions<br />
of deep seasonal melting the dune belt, formed of fine<br />
sands, is subjected to deflation and thermoerosion. The<br />
extent and rate of these processes are so great that in<br />
places the surface of the island became 1-3 m lower than<br />
before the period of exploitation began. Deflation hollows<br />
became widespread. Numerous deflation-thermoerosional<br />
gullies were formed in the coastal cliffs. As a result the<br />
cliffs become lower, their homogeneity had been<br />
disturbed, the coastal zone loses sediments and, finally,<br />
the rate of coastal retreat increases.<br />
During the <strong>20</strong>00 field season we measured the rates of<br />
deflation and thermoerosion with specially equipped<br />
stations. The averaged data of repeated measurements at<br />
more than 50 reference squares has shown that the<br />
thickness of the sand layer blown away by the wind was<br />
IOto 14 cm in technogenically-deformed territories. At the<br />
same time, eolian accumulation took place in the<br />
territories that were not affected by human activity. At the<br />
“erosional” station, we observed the formation of a new<br />
big gully (up to 4 m deep) in the coastal bluff. Up to 400<br />
m3 of sand were removed from the gully itself and from its<br />
catchment area during the two weeks of snow melting in<br />
June. The coast near the Novyi Varandei settlement is a<br />
119
Orgodov, SA.: Coastal dynamics of the Pechora Sea ...<br />
region of wave energy flow divergence and correspon- ACKNOWLEDEMENT<br />
dingly, formation of sediment flows. As a result, coastal<br />
protection in the area close to the Novyi Varandei settle- This work was supported by INTAS grant no. 1489.<br />
ment caused a decrease in sediment supply to the adjacent<br />
areas and, hence, their erosion.<br />
The height of storm surges increased after an earthdam<br />
and a bridge were constructed in the eastern part of<br />
Varandei Island. The latter is an important factor in coastal<br />
dynamics. Previously, during high surges that corresponded<br />
in time with tides, water flowed partly into branches<br />
and channels, thus lowering the surge height and decreasing<br />
its influence upon the coast.<br />
The erosion rate increased considerably under human<br />
impact in the middle-late seventies. In some years it was<br />
up to 7-10 mlyear. It decreased slightly, down to 1.5-2<br />
mlyear, after the coastal protection near the Novyi Varandei<br />
settlement was constructed. However, it remained high<br />
in the adjacent areas. Recent measurements during 1987-<br />
<strong>20</strong>00 have shown that the rate of coastal retreat in the<br />
region, outside of the settlement, have increased and<br />
reached 3-4 mlyear (fig. 2), that is twice as high as in<br />
similar regions that have no human activity.<br />
I^ 01 * I ; “FnM”, ”,<br />
-1 <strong>20</strong> 40 60 80 100 1<strong>20</strong><br />
L, m - dstance +om the base<br />
Figure 2. Dynamics of the coast near Varandei settlement.<br />
The geoecological situation on Varandei Island is<br />
nearly critical. Industrial exploitation of the territory has resulted<br />
in not only the destruction of the natural coastal<br />
system but also caused considerable damage to human<br />
activities. Several housing estates and industrial constructions<br />
have already been destroyed due to coastal retreat.<br />
This damage will increase every year following the cliff retreat<br />
towards the center of Varandei settlement. Oil terminal,<br />
airport and other industrial objects are endangered. It<br />
is therefore necessary to thorough analyze coastal<br />
morpholithodynamic schemes before natural environments<br />
are disturbed.<br />
Active industrial exploitation of the Pechora region demands<br />
a well-thought-out strategy of territory development<br />
and the finding of proper areas for new constructions. This<br />
negative example from the Varandei region shows that a<br />
well-developed ecologically grounded approach to further<br />
exploitation of coastal regions is necessary. After many<br />
years of investigations in the Pechora Sea region, the<br />
<strong>Research</strong> Laboratory of the Geoecology of the North,<br />
MSU, has worked out a special methodology of<br />
morpholithodynamic research and created a database on<br />
the coastal morpholithodynamics of this area that will be a<br />
basis for solving both fundamental and applied problems<br />
arising in the course of coastal reclamation.<br />
1<strong>20</strong>
Permafrost processes inducing disastrous effects in engineering<br />
constructions. Medvezhje gas-field, Western Siberia<br />
Y. V. Panova<br />
Earth Cryosphere Institute, SB RAS, Moscow, 119991, Russia, yanusia@mail.tascom.ru<br />
The Medvezhje gas-field is located in the northern part of<br />
Western Siberia. There are three main gas pipelines constructed<br />
above ground, they have diameters of 14<strong>20</strong> mm<br />
and 10<strong>20</strong> mm. The average gas temperature is more than<br />
15 "C. The permafrost in the area is discontinuously distributed<br />
and consists of sedimentary soil of Quaternary<br />
age (mainly at the Kazanskaya and Salexardskaya sites;<br />
Nedra 1989, Express information 1980). A number of earlier<br />
investigations have been conducted in the area under<br />
study (Galiulin et al. 1997).<br />
The Medvezhje gas pipeline system (MGP) is in need<br />
of reconstruction because of the increasing amount of<br />
deformations that occur. MGP was investigated from the<br />
air and in the field to assess its stability.<br />
Besides the engineering issues, a geocryological map<br />
of 1 :lOO'OOO scale was prepared for the area. Locations of<br />
exploration and the associated findings were marked on<br />
the map and listed in a table. For a small selection see<br />
Table 1. Inspection of the MGP revealed 66 damage sites:<br />
18 were a little deformed, 35 middle deformed and 13<br />
badly deformed.<br />
The most widespread type of deformation (Le. bending<br />
of the pipeline) occurs in a horizontal plane. Usually the<br />
pipeline sags and it results in subsequent ground contact.<br />
An indication for the initiation of pipeline deformation is<br />
the appearance of an apparently superfluous length of<br />
pipeline. The observed pipeline deformations are caused<br />
by different processes (Degtyarev 1974):<br />
temperature deformations;<br />
deformations from inner pressure in the pipeline;<br />
. deformations caused by the incorrect loading of<br />
the pipeline or by other engineering constructions<br />
on and around the MPG;<br />
Deformation or loss of support of the ground underneath<br />
the pipeline also determines the character and degree<br />
of deformation. The latter deformations are influ-<br />
enced by the interaction of the pipeline with permafrost<br />
and geotechnical I geocryological conditions along the<br />
MGP route:<br />
the pipeline embankment or the pipeline itself hinder<br />
surface water runoff;<br />
0 the frozen foundation melts under the thermal influence<br />
of the gas pipeline and the skeleton of the<br />
foundation is compressed;<br />
the pipeline is clamped on watersheds (e.g., hills),<br />
but is without support between such fixed points.<br />
This may lead to differential sagging of the pipeline<br />
under its own weight, and, thus, to incorrect<br />
pipeline loading (Fig. 1).<br />
." ballast-<br />
Figure 1. Schematic of an incorrectly ballasted pipeline.<br />
Various combinations of the above listed processes<br />
have been found.<br />
Deformations, caused by the wrong loading of the<br />
pipeline due to the influence of additional engineering<br />
constructions are:<br />
destruction of the MGP embankment and the subsequent<br />
relocation of the embankment ground;<br />
. part of the pipeline is compressed by the superimposed<br />
road embankment that crosses it (Fig. 2).<br />
Road embankment<br />
Figure 2. Schematic image of the vertical deformation<br />
pipeline as it crosses a highway.<br />
of the gas-<br />
After inspection of the MGP and analysis of the field<br />
data we marked out the general exogenous and geologi-<br />
121
cal processes which cause changes in the MPG location<br />
and result in its deformation.<br />
During the operation of the MGP the bog zones in the<br />
area were found to increase under the thermal influence of<br />
the pipeline, as has been confirmed by air-photos over the<br />
years. If the foundation ground is ice-rich and able to subside,<br />
water often initiates thermokarst processes: Particularly<br />
bog water, temporary or constant brooks, subsurface<br />
water, within or above the active layer, influence the pipeline.<br />
In a number of cases drainage was found to occur<br />
lengthwise along the pipeline embankment, for instance,<br />
in the line trench. For elevated sites, the concentration of<br />
water flow in this trench leads to water inundation into the<br />
rock.<br />
At the beginning of gas extraction holes began to form<br />
in the foundations, by thawing, because of missing heatinsulation.<br />
On ice-rich sites, the thawing was accompanied<br />
by the formation of thermokarst slumps. The most typical<br />
types of subsequent deformations are:<br />
non-uniform pipeline slumps accompanied by bends in<br />
the vertical and horizontal planes;<br />
emergence of the pipeline from water, accompanied by<br />
bends in the vertical and horizontal planes.<br />
The pipeline, initially and most frequently, was built on<br />
ice-rich highland, 2-3 m above the ground. Initially it contained<br />
vertical bends, they slowly straightened due to the<br />
ice-rich ground settling because of ice melt.<br />
Beside the thermokarst processes within the slopes of<br />
a river or brook, water erosion also plays a part in the formation<br />
of settlements.<br />
An enhanced deformation of the MPG occurs at permafrost<br />
sites containing ice wedges due to the melting of<br />
the ice bodies. Investigated sites that developed such<br />
Panova, V.V,: Permafrost processes inducing disastrous effects<br />
thermokarst phenomena are confined to polygonal peat<br />
soil with ice wedges or with ice inclusions. In extensive<br />
peat soil areas the pipeline comes out, in general, into the<br />
drained trench which is supported by local peat-material or<br />
by mineral ground with only a thin layer of peat. Many<br />
sites are deprived of an embankment and the open trench<br />
forms into a pond which becomes a thermokarst center<br />
along the ice wedge. Heat transferred by water-flow forms<br />
an initial trench in the ice wedges. They increasingly melt<br />
forming growing trenches filled with water.<br />
The pipeline itself is a powerful heat source, especially<br />
in the summer time. A broken embankment leads to a<br />
change in albedo and thus, to ground warming. As a result<br />
of the deformation and destruction of the melting peat at<br />
the base it is possible for the pipeline to sink to the foundation<br />
ground owing to the thermal influence of the trench<br />
water on the peat soil, even if the ground contains no ice.<br />
This effect lasts until all the peat from the pipeline base<br />
has been pushed away and firm mineral soil has been<br />
reached.<br />
REFERENCES<br />
”Geocryology of USSR Western Siberia” “Nedra” Moscow 1989.<br />
Degtyarev, 1974. “Instruction on account and choice of designs of<br />
holev with thermal protection in a permafrost” Moscow “VSRlgas”<br />
72.<br />
Galiulin f., lsmailov I., Dubinin P. and Samsonova L.N. 1997. Increase<br />
of efficiency of development of gas deposits of Far North<br />
Moscow “Science” 655<br />
Express information, release 3, Moscow. 1980. Gas industry series:<br />
Geology, drilling and development of gas deposits.<br />
Tablel. Example of MGP data taken in file research, summer 1999<br />
Site No. Geomor- Length (L), amplitude (A) and Filling up the Type of construction method and Degree of<br />
phologic type of MGP<br />
ponds with water preservation of the embankment deformation<br />
type of deformation (m)<br />
round the pipeline<br />
site in horizontal plane in vertical<br />
plane<br />
(m)<br />
flood- -11- there is above ground, middle de<br />
46-2 plain of<br />
destroyed by water embankment<br />
brook<br />
504-1 bog landscape<br />
right bend A-<strong>25</strong>; -11-<br />
left bend A-2<br />
there is above ground, strongly d<br />
destroyed by water embankment<br />
122
8th International conference an Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Frozen ground data information: advances in the IPA Global<br />
Geocryological Data (GGD) system<br />
M. A. Parsons, T. Zhang, R. G. Barry and *J. Brown<br />
National Snow and Ice Data CenferlWorld Data Center for Glaciology, Boulder, CO, USA<br />
* lnfernafional Permafrost Association, Woods Hole, MA, USA<br />
BACKGROUND<br />
Permafrost regions occupy about 23.9% and seasonally<br />
frozen ground regions underlie on average of approximately<br />
57.1 % of the exposed land surface in the Northern<br />
Hemisphere (Zhang et al. <strong>20</strong>03). Data and information on<br />
frozen ground collected over many decades and in the<br />
future are critical for fundamental process understanding,<br />
environmental change detection and impact assessment,<br />
model validation, and engineering application in seasonal<br />
frost and permafrost regions (IPA DlWG 1998). However,<br />
many of these data sets and information remain widely<br />
dispersed and relatively unavailable to the national and<br />
international science and engineering community, and<br />
some are in danger of being lost permanently.<br />
The International Permafrost Association (IPA) has<br />
long recognized the inherent and lasting value of data and<br />
information and has worked to prioritize and assess frozen<br />
ground data requirements and to identify critical data sets<br />
for scientific and engineering purposes. The groundwork<br />
of the IPA strategy for data and information management<br />
was laid during the 1988 IPA Conference in Trondheim,<br />
Norway. Barry (1988) presented a paper on status and<br />
needs of permafrost data and information and organized<br />
an informal workshop on requirements for permafrost data<br />
and information that attracted more than 100 attendees.<br />
The IPA Data and Information Working Group (DIWG)<br />
was established and a formal workshop was held during<br />
the 1993 IPA Conference in Beijing, China. The DlWG has<br />
subsequently organized workshops in Southampton and<br />
Oslo (1994), Potsdam (1995), and Boulder (1996). The<br />
Global Geocryological Data (GGD) project was initiated by<br />
the DlWG during the Boulder workshop in 1996. During<br />
the 1998 IPA Conference in Yellowknife, Clark and Barry<br />
(1998) provided an update on IPA permafrost data and<br />
information management. The first Circumpolar Active<br />
Layer Permafrost System (CAPS) CD-ROM was published<br />
and delivered to the delegates. The IPA Standing<br />
Committee on Data, Information and Communication<br />
(SCDIC) was established, replacing the former IPA DIWG.<br />
FROZEN GROUND DATA CENTER (FGDC)<br />
To continue the IPA strategy for data and information<br />
management and to meet the requirements by cold regions<br />
science, engineering, and modeling community, the<br />
World Data Center (WDC) for Glaciology, Boulder in collaboration<br />
with the International Arctic <strong>Research</strong> Center<br />
(<strong>IARC</strong>) has established a new Frozen Ground Data Center<br />
(FGDC) as a key node in the IPAs GGD system. The IPA<br />
SCDIC has organized two workshops in Boulder (<strong>20</strong>02<br />
and <strong>20</strong>03) on the prioritization and assessment of frozen<br />
ground data requirements and identification of data sets.<br />
The FGDC has expanded access to the 1998 CAPS data<br />
and has augmented data holdings. Data and information<br />
for seasonally frozen ground regions from in situ measurements,<br />
satellite remote sensing, and model outputs are<br />
included. Figure 1 is an example of these new types of<br />
data. It shows modeled maximum frozen ground depth for<br />
non-permafrost regions in the Arctic drainage basin (Oelke<br />
et al. <strong>20</strong>03).<br />
Figure 1. Maximum frozen ground depth 1998/99<br />
D<br />
<strong>25</strong>0<br />
2<strong>25</strong><br />
<strong>20</strong>0<br />
175<br />
150<br />
1<strong>25</strong><br />
100<br />
75<br />
50<br />
<strong>25</strong><br />
0<br />
[cm]<br />
Rmax<br />
The FGDC is distributing a new version of CAPS at the<br />
<strong>July</strong> <strong>20</strong>03 International Conference on Permafrost in <strong>Zurich</strong>.<br />
The ultimate goal of the FGDC is to serve as a central<br />
access point for frozen ground data and information in<br />
123
the international permafrost community and scientific<br />
community at large.<br />
The FGDC has improved access to existing data<br />
through an online search and order system and availability<br />
in the Global Change Master Directory. The FGDC has<br />
also expanded and updated current holdings with maps of<br />
global and regional permafrost, soil temperature, and soil<br />
classification in a variety of grids and data formats especially<br />
geared to aid the permafrost modeling community.<br />
Publication and distribution of CAPS version 2 marks a<br />
milestone for the FGDC and the IPA. CAPS version 2 is a<br />
compendium of GGD data and metadata currently available<br />
in the FGDC and around the world (see Table 1). It<br />
serves as a snapshot of frozen ground and related data<br />
holdings available as of spring <strong>20</strong>03 and enables access<br />
to these data for users lacking ready internet access or<br />
high-speed connections.<br />
Table 1. Highlights of CAPS version 2<br />
IPA Program Data<br />
Metadata from the Global Terrestrial Network for Permafrost's<br />
(GTN-P) Borehole Program and the Permafrost and Climate<br />
in Europe (PACE) project.<br />
Annual thaw depth and other summary data from the GTN-P's<br />
Circumpolar Active Layer Monitoring (CALM) Program.<br />
IPA Cryosols Working Group map of Arctic soils<br />
Metadata, photos, and a specialized bibliography from the<br />
Arctic Coastal Dynamics (ACD) project<br />
Review papers and a specialized bibliography form the<br />
Southern Hemisphere Working Group (SHWG)<br />
Maps<br />
The IPA Circum-Arctic Map of Permafrost and Ground Ice in<br />
several new grids and formats.<br />
Regional permafrost maps from Russia, China, Europe, and<br />
Alaska<br />
Northern Circumpolar Soils Map<br />
Regional soil maps from Russia and China<br />
Frozen Ground<br />
Historical soil temperatures from stations across Russia<br />
Soil temperatures from sites in Alaska, China, Mongolia, and<br />
Antarctica.<br />
Global grids of seasonal frozen ground derived from remote<br />
sensing<br />
Borehole and active layer data from historical sites and other<br />
sites outside the GTN-P purview.<br />
Modelled seasonal frozen ground depth for the Arctic<br />
drainage basin.<br />
Bibliographies<br />
A Comprehensive Permafrost Bibliography current through<br />
Decemeber <strong>20</strong>02. (A searchable, up-to-date version is<br />
available online at the FGDC web site.)<br />
A bibligraphy of permafrost and related maps<br />
Specialized bibliographies from the SHWG and ACD.<br />
Other Products<br />
Output from several permafrost or active layer models.<br />
Borehole and active layer data from historical sites and other<br />
sites outside the GTN-P purview.<br />
Dozens of regional and site-specific data sets of permafrost,<br />
soil properties, snow cover, and related parameters.<br />
GGD metadata describing over 100 other data sets available<br />
from investigators and institutions around the world.<br />
Identification of additional data and information from all<br />
participating countries, organizations, and individuals are<br />
requested. Suggestions on data acquisition, management<br />
and distribution are always welcome and encouraged. The<br />
IPA Standing Committee on Data, Information and Communication<br />
will continue to coordinate the GGD and CAPS<br />
activities.<br />
ACKNOWLEDGEMENTS<br />
We thank the IPA Standing Committee on Data, Information<br />
and Communication for coordinating the GGD and<br />
CAPS activities. We especially thank the many data contributors<br />
to this important effort. This work was supported<br />
by the International Arctic <strong>Research</strong> Center, University of<br />
Alaska Fairbanks, under the U.S. NSF cooperative agreement<br />
OPP-0002239. Financial support does not constitute<br />
endorsement of the views expressed in this article.<br />
REFERENCES<br />
Barry, R. G. 1988. Permafrost data and information: status and needs.<br />
In Senneset, K. (ed.), <strong>Proceedings</strong> of the 5th International Conference<br />
on Permafrost, Trondheim, Norway, 1988. Vol. 1, Tapir Publishes,<br />
Trondheim: 119-122.<br />
Clark, M. J. and Barry, R. G. 1998. Permafrost data and information:<br />
advances since the Fifth International Conference on Permafrost,<br />
in Lewkowict, A. and M. Allard (eds.), <strong>Proceedings</strong> of the 7th International<br />
Conference on Permafrost, Yellowknife, Canada,<br />
1998: 181-188.<br />
International Permafrost Association, Data and Information Working<br />
Group, comp. 1998. An Introduction to Circumpolar Active-Layer<br />
Permafrost System (CAPS), in Circumpolar Active-Layer Permafrost<br />
System (CAPS), version 1.0. CD-ROM available from National<br />
Snow and Ice Data Center, Boulder, Colorado.<br />
Oelke, C., Zhang, T., Serrere, M. and Armstrong, R. in press. Regional-scale<br />
modeling of soil freezelthaw over the Arctic drainage<br />
basin. Journal of Geophysical <strong>Research</strong>.<br />
Zhang, T., R., Barry, G., Knowles, K., Ling, F. and Armstrong, R. L.<br />
<strong>20</strong>03. Distribution of seasonally and perennially frozen ground in<br />
the Northern Hemisphere. <strong>Proceedings</strong> of the Eighth International<br />
Conference on Permafrost, <strong>Zurich</strong>, <strong>Switzerland</strong>, <strong>July</strong> 21-<strong>25</strong>, <strong>20</strong>03.<br />
We continue to welcome data submissions for inclusion<br />
on the FGDC web site at http://nsidc.org/fgdc/index.html.
8th International Conference on Permafrost Extended Abstracts an Current <strong>Research</strong> and Newly Available ~ ~ ~<br />
Mapping of rock glaciers with optical satellite imagery<br />
F. Paul, A. Kaab and W. Haeberli<br />
Glaciology and Geomorphodynamics Group, Department of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
There is a growing concern about permafrost degradation<br />
as a result of ongoing global warming. In particular, rock<br />
fall events and debris flows from now unstable mountain<br />
slopes are expected (Haeberli and Beniston 1998).<br />
Hence, the great effort to map the distribution of mountain<br />
permafrost in the Alps. Rock glaciers are indicative of<br />
permafrost conditions and can easily be recognized on<br />
aerial photography due to their typical shape. Time series<br />
analysis of the derived orthophotos allows the calculation<br />
of flow fields and mass changes (Kaab <strong>20</strong>02) but is not<br />
practical for large area mapping on a small scale.<br />
The potential of high-resolution panchromatic satellite<br />
sensors to detect rock glaciers has been analysed in order<br />
to validate the results of permafrost distribution models at<br />
large scales and in remote areas. The investigated sensors<br />
range from I5 m spatial resolution ETM+ to the 1 m<br />
lkonos sensor. The sensors also differ in the spectral<br />
range and ground area covered as well as costs per km2<br />
(Table 1). In general costs increase with higher resolution<br />
while the area covered decreases.<br />
Table 1 Some characteristics of the pan sensors examined.<br />
-<br />
ETM+ SPOT IRS-IC lkonos<br />
Resolution [m] 15 10 5 1<br />
Spectral range [mm] 0.52-0.9 0.51-0.73 0.5-0.75 0.45-0.9<br />
Area covered [km] 180x180 60x60 70x70 11x1 00<br />
Costs per km2 [US$] 0.02 0.33 0.51 ca. 30<br />
In Figure 1 the comparison of the three pan bands from<br />
ETM+, SPOT and IRS-IC is shown together with a ground<br />
based photo of a complex rock glacier system near Piz<br />
d’Err (Crisons). The ETM+ sensor (Fig. la) captures the<br />
rock glaciers quite well as the sensor is also sensitive in<br />
the near infrared where the high reflection of vegetation<br />
contrasts well with the low reflection of bare rock. However,<br />
details of the ridge structure are not visible and the<br />
spatial resolution is thus, not appropriate for a clear identification.<br />
In the SPOT image (Fig. I b) the typical ridges become<br />
visible and identification and large area mapping of rock<br />
glaciers seems feasible. However, the outline of the lower<br />
(relict) part is difficult to recognize as the reflections from<br />
the surrounding meadow and bare rock are quite similar in<br />
this spectral range.<br />
The IRS-IC image (Fig. IC) reveals only few new details,<br />
as compared to SPOT, but seems to be more noisy<br />
(nominal resolution is 6.8 m). The surrounding meadow is<br />
slightly brighter than seen with SPOT although the spectral<br />
range covered is quite similar.<br />
A comparison between SPOT and lkonos imagery<br />
(draped over a DEM with <strong>25</strong> m spatial resolution) is shown<br />
in Figure 2 (view to south) for the comparable small Gigli<br />
rock glacier (near Sustenpass, Uri). Although some of the<br />
rock glacier ridges are visible in the SPOT imagery (see<br />
inset), it is by no means comparable to the level of detail<br />
visible from Ikonos. Here also, the typical larger rocks at<br />
the rock glacier surface and the very fine debris at the<br />
front can be discerned. A DEM allows the creation of 3D<br />
perspective views which enhance the identification of further<br />
rock glaciers (white arrows). A large number of<br />
breaches from former debris flows across historic (1850)<br />
moraines can be identified as well.<br />
The regions of likely and probable permafrost are indicated<br />
as high-lighted areas in Figure 2. The model simu-<br />
lates BTS, a threshold of -2.0(-3.0)<br />
OC is chosen to assign<br />
regions with likely (probable) permafrost (Gruber and<br />
Hoelzle <strong>20</strong>01). It seems that the rock glaciers originate in<br />
regions with likely permafrost but extend much more further<br />
down into permafrost free terrain. This has also been<br />
observed for many other rock glaciers (Frauenfelder et al.<br />
<strong>20</strong>01).<br />
We propose that contrast enhanced SPOT pan imagery<br />
can be used to identify and map rock glaciers for<br />
large and remote areas at comparably low costs. The IO<br />
m spatial resolution seems appropriate for validation of<br />
permafrost distribution maps obtained from <strong>25</strong> m spatial<br />
resolution DEM data. For further detailed studies of interesting<br />
regions the I m resolution lkonos imagery is a<br />
valuable alternative to aerial photography but about three<br />
times more expensive.<br />
1<strong>25</strong>
Paul, F. et. al.: Mapping of rock glaciers with optical satellite imagery<br />
REFERENCES<br />
Frauenfelder R, Haeberli W, Hoelzle M. and Maisch M. <strong>20</strong>01. Using motely sensed data. Permafrost and Periglacial Processes 12 (I):<br />
relict rockglaciers in GIS-based modelling to reconstruct Younger 69-77.<br />
Dryas permafrost distribution patterns in the Err-Julier area, Swiss Kaab, A. <strong>20</strong>02. Monitoring high-mountain terrain deformation from re-<br />
Alps. Norwegian Journal of Geogra-phy 55 (4): 195-<strong>20</strong>2.<br />
peated air- and spaceborne optical data: examples using digital<br />
Haeberli, W. and Beniston, M. 1998. Climate change and its impacts aerial imagery and ASTER data. ISPRS Journal of Photogramon<br />
glaciers and permafrost in the Alps. Ambio 27 (4): <strong>25</strong>8-265. metry and Remote Sensing 57 (112): 39-52.<br />
Gruber, S. and Hoelzle, M. <strong>20</strong>01. Statistical modelling of mountain<br />
permafrost distribution: local calibration and incorporation of re-<br />
126
Morphogenetic classification of the Arctic coastal seabed<br />
Y. Pavlidis and S. Nikiforov and *V. Rachold<br />
P.P. Shirshov Institute of Oceanology, Moscow, Russia<br />
*Alfred Wegener Institute for Polar and Marine <strong>Research</strong>, Pofsdam, Germany<br />
INTRODUCTION<br />
A precise and standardized classification on the variety of relief<br />
forms of the Arctic shelf is essential for both scientific and<br />
applied purposes. The evolution of the Arctic coastal zone,<br />
which is controlled by the interactions of modern wave and<br />
ice factors and the influence of numerous glaciations and<br />
large-scale sea level fluctuations, significantly differs from<br />
other coastal areas of the World Ocean. Glaciations have left<br />
coastal traces in the western part of the Russian Arctic,<br />
whereas its eastem part was almost completely drained during<br />
glacial periods and the subaerial paleorelief is characterized<br />
especially by cryogenic forms: thermokarst, theme<br />
abrasion, thermodenudation and ctyodislocation.<br />
Earlier classifications of the coastal seabed relief suggested<br />
by lonin et al. (1990) for the World Ocean or by<br />
Shepard (1977) according to the peculiarity of geological<br />
structures or sedimentation processes could not characterize<br />
the variety of forms and regional features in the Arctic zone.<br />
The classification presented in this paper is based upon scientifically<br />
evidenced information on morphology, origin and<br />
age as well as geological and neotectonic characteristics d<br />
the seabed relief. This kind of simultaneous analysis of complex<br />
of interactive factors can be called a “morphogenetic”<br />
approach.<br />
The investigations are based on the results of long-term<br />
research with the application of modern equipment including<br />
narrow-beam and multi-beam echo sounders, Parasound,<br />
side-looking radars, bathymetric maps and the analysis d<br />
numerous sediment samples. According to the Science and<br />
Implementation Plan of the Arctic Coastal Dynamics (ACD)<br />
project (IASC <strong>20</strong>01), the classification will be formulated for<br />
integration into a circum-Arctic coastal Geo Information System<br />
(GIs).<br />
RESULTS AND DISCUSSION<br />
The seabed evolution is determined by both modem and<br />
paleogeographical processes such as vertical tectonic and<br />
glacio-isostatic movements. Among the exogenic and en<br />
dogenic coastal processes it is possible to allocate active<br />
processes, which directly contribute to the formation of the<br />
shelf relief, and passive processes, which predetermine the<br />
display of the active processes and direct the course of their<br />
development. Active processes can be modern (hydrogenic,<br />
gravitational, etc.), paleogeographical (glacial, erosional, etc.),<br />
and in some cases anthropogenic processes. Structural<br />
forms, with few exceptions, are related to passive morphogenetic<br />
factors. By using such an approach it is possible b<br />
designate three major categories of the seabed relief: structural,<br />
structural-sculptural and sculptural. This kind of subdivision<br />
is relative in many respects because the majority of relief<br />
forms are caused by both endogenic and exogenic facto<br />
andprocesses. It is necessary todeterminetheprevailing<br />
value of various factors of concrete seabed forms. A certain<br />
cannot be avoided subjectivity in allocating structural and<br />
sculptural forms of a relief, and in their morphogenetic division<br />
into active and passive factors. The need to characterize the<br />
relief “layer by layer” arises inevitably when specialized<br />
geomorphological maps have to be produced and a database<br />
to be used in a GIS environment within in international network<br />
has to be designed. As an example the following layers<br />
could be used: first layer - endogenic background; second<br />
layer - relief forms linked to the action of paleogeographic<br />
factors; third layer - modern relief forms caused by the action<br />
of endogenic processes.<br />
Typical structural forms are large seabeds predetermined<br />
by geological structures and created by endogenic<br />
processes, Le., folded andlor fault-fracture tectonics,<br />
volcanism, vertical movements of the Earth’s crust, etc.<br />
Geological structures underlie all forms of the shelf relief and<br />
determine the position of large relief forms, such as synclinal<br />
underwater depressions, anticlinal and brachyanticlinal<br />
underwater raisings, monoclinal plains, flexure ledges etc.<br />
This large group can be further divided into two genetic types<br />
and some subtypes. (1) The tectogenic type includes plicated<br />
forms, Le., anticlinal, brachyanticlinal, synclinal, brachysynclinal,<br />
monoclinal, uniclinal; disjunctive, i.e., fault and<br />
grabens, horsts, fracture-blocks; and non-broken structural<br />
dislocations, i.e., subhorizontal and inclined plains. (2) The<br />
volcanogenic type includes magmatic and mud volcanic<br />
subtypes. Usually relief forms created by volcanic<br />
processes like lava domes and mud volcanic cones<br />
127
Pavlidis. Y. et al.: Arctic coastal seabed..<br />
(Norwegian Sea) are located in water depthsofmorethan<br />
<strong>20</strong>0 m.<br />
In our morphogenetic classification, structural-sculptural<br />
forms are assigned to a category of transitive, ofien relict formations.<br />
These forms can be divided into the following sub<br />
types: (1) erosion-fectonic, Le. , structural-erosive with fluvial<br />
or glacial relief, structural-erosive-gravity created by turbidity<br />
currents, (2) glacial-tectonic, ;.e., structural depressions in<br />
internal basins of fjords and covered by glacial-marine deposits,<br />
structural swells with accumulative superstructure, structural<br />
forms with a surface of glacial ablation corresponding ID<br />
fjords and adjacent underwater trough valleys, structural<br />
forms with glacial ablation and accumulative marginal troughs<br />
and anticlinal swells blocked by moraines.<br />
Exogenic subaerial and subaquatic processes are the<br />
most active ones. The sculptural relief of the seabed is created<br />
by destructive and accumulative processes such as<br />
ablative and accumulative activity of Late Quaternary glaciers<br />
on the shelves of the marginal and internal seas of polar<br />
and subpolar climatic zones. Glacial forms are most typical<br />
amongtherelictexogenicsculpturalreliefformswithin the<br />
limits of the glacial shelf in the Arctic. The sculptural relief<br />
forms can be divided into seven types and several subtypes:<br />
This study was supported by INTAS (project number <strong>20</strong>01-<br />
2332).<br />
(I) glaciogenic, Le., glacial-ablative, glacial-accumulative and<br />
glacial-dynamic, (2) cryogenic, ;.e., thermokarst relict forms<br />
such as extended underwater depressions and hollows REFERENCES<br />
documented by echosounding and acoustic profiling in the<br />
Laptev, East Siberian and Pechora Seas; the origin of these IASC <strong>20</strong>01. Arctic Coastal Dynamics (ACD). Science and Implenegative<br />
relief forms is connected with the thermal destruction mentation Plan, International Arctic Science Committee, Oslo,<br />
April <strong>20</strong>01.<br />
of ice lenses and wedges during the late glacial transgreslonin,<br />
A., Pavlidis Yu. and Yurkevich, M. 1990. Relief of arctic eastsion,<br />
the size ranges from 2 to IOm depth, 10 to I00 m width ern shelf of Russia and its classification. In A Aksenov (eds),<br />
and extension of 3 to 4 km depending on the thickness of the Geology and geomorphology of shelf and continental slope.<br />
lenses of repeatedly-cavern-load ice; in the Chukchi Sea, the Moscow, Nauka: 24-50.<br />
relict thermokarst relief forms are large depressions which are Shepard, F.P. 1977. Geological oceanography. N.Y.<br />
covered by a 7 to IOm thick Holocene sediment layer;<br />
cryodislocative relief forms represented by folds and frost<br />
mounds have a local distribution in the Arctic, for example<br />
Western Yamal Peninsula, (3) hydrodynamic, ;.e., wave<br />
abrasion forming modem cliffs and benches and paleo underwater<br />
terraces, wave accumulation with ancient underwater<br />
accumulative forms and coastal bamer islands, spits,<br />
beaches, etc., abrasive-accumulative processes forming<br />
modem underwater coastal slopes, mainly subglacial accumulative<br />
and tidal forms such as sandy waves and ridges,<br />
tidal troughs, step tidal benches, etc., (4) torrentogenic, Le.,<br />
accumulative and erosive as, for example, in the Bering<br />
Strait where, due to strong quasistationary currents, a subhorizontal<br />
plain lacking modern deposits is formed in the central<br />
zone while as a result of deposition, underwater cones<br />
are generated in the adjoining southem Chuchi Sea, (5) fluvial,<br />
ire., fluvio-glacial and potamogenic forms with icemarginal<br />
valleys andchannelsofpaleo-rivers,ancientand<br />
modem deltas, river mouth bars, etc., (6) gravitational, Le.,<br />
rockfall-landslides, turbidity currents, and (7) modem anthropogenic,<br />
Le., constructive forms connected with ground spoil,<br />
artificial islands, basements of port constructions, oil derricks<br />
etc., destructive forms with artificial cuts, port channels, waterways<br />
in gulfs and other near-port constructions within<br />
shallow water.<br />
CONCLUSIONS<br />
The Arctic shelf can be described as a zone witch evolved in<br />
the come of long-term endogenic and exogenic interactions.<br />
The initial structure was created by the basement surface<br />
which was and is constantly reworked by a complex of paleogeographical<br />
and modem processes within various glacial<br />
and interglacial periods and sea level fluctuations. The classification<br />
suggested in the present article is based on scientifically<br />
evidenced data on morphology, origin and age, and on<br />
the geological and neotectonic features of the seabed relief,<br />
and it also takes into account a multitude of natural factors and<br />
their interactions and evolution in time. Three types of major<br />
relief-forming processes were distinguished and further subdivided:<br />
structural, structural-sculptural (transitive), and sculptural<br />
(formed by modem and paleogeographical exogenic<br />
processes).<br />
128<br />
ACKNOWLEDEMENT
Near-surFace ground temperatures and permafrost distribution at<br />
Gornergrat, Matter Valley, Swiss Alps<br />
S. Philippi, T. Hen and 1. King<br />
Institute of Geography, Justus Liebig University, Giessen, Germany<br />
INTRODUCTION<br />
Due to substantial research during the past fifteen years,<br />
permafrostdistributionoftheMatter valley is known on an<br />
overview scale. However, the currently available permafrost<br />
distribution models do not consider local soil properties which<br />
can vary significantly within short distances in high mountain<br />
environments. Besides other local factors such as air<br />
temperature, solar radiation and snow cover, changes in<br />
substrate and moisture conditions strongly influence the<br />
ground thermal properties (Williams and Smith 1989) and<br />
consequently, the distribution pattern of mountain permafrost.<br />
The zone of discontinuous permafrost is characterized by<br />
rather warm ground temperatures insignificantly below zero,<br />
which makes permafrost at its lower limit very susceptible b<br />
increasing air temperatures. Active layer thickening, rising d<br />
the lower limits of permafrost and permafrost reduction by<br />
basal melting could be possible responses to long-term<br />
climate changes (Williams and Smith 1989).<br />
The current investigations will provide quantitative insights<br />
into ground thermal regimes and will correlate measured soil<br />
temperatures with different soil properties and their possible<br />
effects on the discontinuous permafrost pattem.<br />
The Gomergrat area is located between 2700 and 3<strong>20</strong>0<br />
meters a.s.1. at the end of the Matter valley surrounded by<br />
some of the highest peaks of the Alps. The continental<br />
climatic conditions cause a high glacier equilibrium line<br />
associated with a periglacial belt of large vertical extent.<br />
According to King (1996), discontinuous permafrost can be<br />
expected in an altitudinal range between 2800 and 3500<br />
meters a.s.1. and is indicated by the presence of numerous<br />
active rock glaciers.<br />
SELECTED STUDY SITES<br />
The east-west running crest between Gomergrat (3135<br />
meters a.s.1.) and HohGlli (3286 meters a.s.1.) divides the<br />
study area into a northern part called ‘Kelle’ and a southern<br />
part named ‘Tuff. Both areas are located within the zone d<br />
discontinuous permafrost showing a rich periglacial<br />
morphology such as solifluction lobes, patterned ground and<br />
rock glaciers. However, the local permafrost pattem is not<br />
yet known.<br />
Shallow ground temperature measurements are performed<br />
at eight vertical profiles differing in substrate (cf. Table 1) and<br />
moisture conditions. The measurements are carried out in<br />
order to obtain an insight into the various reasons for the<br />
variability of ground thermal properties over short distances<br />
within the zone of discontinuous permafrost.<br />
In terms of local soil characteristics, each of the four<br />
instrumented profiles in the Tuft corresponds to a profile in the<br />
Kelle. The substrates of location T (Tuft) I and K (Kelle) I<br />
show a rather high soil moisture content, T2 and K2 were<br />
installed in substrates with moderate water content, whereas<br />
T3 and K3 are comparatively dry locations. All sites show a<br />
sparse vegetation cover of moss, lichen and grass and are<br />
located at approax 3000 meters ad. (cf. Table I).<br />
According to Harris and Pedersen (1998) the differing<br />
thermal properties of blocky materials seem to explain<br />
permafrost occurrences below the regional lower limit. In<br />
order to comparethethermalregimesofsoilsandcoarse<br />
debris two additional locations (K4 and T4) within the surface<br />
layer of active rock glaciers were chosen.<br />
Meteorological data are available for the ‘Kelle’ and the<br />
permafrost monitoring site at Stockhorn plateau (3410 meters<br />
a.s.1.) close to the research area.<br />
GROUND TEMPERATURE MEASUREMENTS<br />
Measurement setup<br />
In August <strong>20</strong>02 eight data loggers with an accuracy of 0.02<br />
“C were installed, measuring shallow ground temperatures<br />
dierent levels (cf. Table 1). The measurement interval was<br />
set at 15 minutes during summer and autumn, while from<br />
October on hourly recordings are made. So far, data are<br />
availablefromtheendofAugustuntiltheend of October<br />
<strong>20</strong>02.<br />
129
Philippi, S. et al. 1 Near-surface ground temperatures and permafrost ...<br />
Table I. Information on the location of near-surface ground measurements at Gornergrat.<br />
Profile Altitude depth of sensors location soil type humus within<br />
(until 100 cm)<br />
upper horizon<br />
m cm cm %<br />
TI 3005 2, <strong>25</strong>, 50, 100, 160 moist plain 0-15 loam 9.7<br />
15-100 sandy loam<br />
T2 3000 2, <strong>25</strong>, 50, 100, 180 solifluction lobe 0-100 loam 15.0<br />
T3 3000 2, <strong>25</strong>, 50, 100, 150 knoll 0-100 sandy loam 4.5<br />
T4 3010 50,1<strong>20</strong>,190,<strong>25</strong>5 active rock glacier<br />
K1 2950 2, <strong>25</strong>, 50, 100, 150 moist plain 0-1 5 loam 9.5<br />
15-100 silt loam<br />
K2 2950 2,<strong>20</strong>,45,70,1<strong>20</strong> plain 0- 40 sand 2.3<br />
40-100 loamy sand<br />
K3 3005 2, <strong>25</strong>, 50, 100, 140 knoll 0-1 00 sandy loam 6.3<br />
K4 2950 100, 160, 2<strong>20</strong>, 280,360 active rock glacier<br />
P reliminary results<br />
The observed smaller temperature ranges at the moist profiles OUTLOOK<br />
K1 and TI compared to the drier locations, could be<br />
explained by the low thermal conductivity of water. This To verify the local permafrost pattern, measurements of the<br />
effect is shown in Figure 1, comparing the measured ground Basal Temperature of Snow cover (BTS) and further soil<br />
temperatures at the northern site KI (moist) with the south- temperature readings will becarried out before the I COP<br />
exposed site 13 (dry).<br />
<strong>20</strong>03 and will be discussed in the poster presentation.<br />
The study area will be visited during one of the postconference<br />
excursions in <strong>July</strong> <strong>20</strong>03.<br />
REFERENCES<br />
12.Y 13.Y 14.Y 15.Y l6.l 17.9 18.9 199 <strong>20</strong>.4 2l.q 22.9 23.9 24.9 <strong>25</strong>.9 26Y<br />
h i e<br />
Figure 1. Near-surface ground temperatures in 2 cm and <strong>25</strong> cm<br />
depth in the profiles K1 and K2 (12.09.-26.09.<strong>20</strong>02).<br />
Harris, SA. and Pedersen D.E. 1998. Thermal Regimes Beneath<br />
Coarse Blocky Materials. Permafrost and Periglacial Processes<br />
9: 107-1<strong>20</strong>.<br />
King, L. 1996. Dauerfrostboden im Gebiet Zermatt - Gornergrat -<br />
Stockhorn: Verbreitung und permafrostbezogene Erschliessungsarbeiten.<br />
Z. Geomorph. N.F., Suppl. Bd. 104: 73-93.<br />
Williams, P.J. and Smith, M.W. 1989. The Frozen Earth. Cambridge:<br />
University Press.<br />
Differences in aspect due to solar radiation mainly affect<br />
the surface ground temperatures (cf. Figure 1). W&<br />
increasing depth local soil properties - especially moisture<br />
conditions I seem to have a major impact on the thermal<br />
regime. Apart from the surface ground temperatures, which<br />
are considerably higher at the southern location, the two moist<br />
profiles show similar temperatures and temperature ranges at<br />
all other measured levels - independent of their aspect.<br />
The temperature amplitude at all measured levels is<br />
significantly greater within the rock glaciers. The phase lag<br />
between surface and soil temperature changes increasing<br />
with depth is missing at the rock glacier profiles. Similar the to<br />
soil locations, the coarse blocky layers also show a<br />
decreasing amplitude with depth, but without any time delay.<br />
Rapid air movement through the gaps seem to be the main<br />
process of heat transfer (Harris and Pedersen 1998) and<br />
would explain the direct response of the temperature changes<br />
within the block layers to a change in air temperatures at the<br />
surface.<br />
130
8th International Conferen~e an Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available information<br />
Results of the engineering geocryological mapping of the Pechora Sea<br />
arctic coast (the key site of the Varandey Peninsula)<br />
A. A . Popova<br />
Industrial and <strong>Research</strong> lnstifute for Engineering of Construction, Moscow, Russia<br />
INTRODUCTION AND STUDIED AREA<br />
Natural conditions are complex in Arctic Sea coastal areas.<br />
At present, it is necessary to study this area in detail in connection<br />
with oil and gas recovery and all the associated<br />
pipework.<br />
The studies were performed in the Pechora Sea coastal<br />
band with a width of 8-10 km: from the Varandey Cape in the<br />
west to the Khaipudyr Bay in the east. The average annual<br />
air temperature is -5.6 'C. The studies were performed on<br />
the laida (absolute elevation 0-3 m) and the first sea terrace<br />
(absolute elevation 515 m). The sea terrace surface is complicated<br />
by numerous shallow lakes and sedge-moss bogs.<br />
Laida is a low part of the coast and subject to flooding during<br />
tides and storms. Saline sea water penetrates, through the<br />
river valleys, up to IOkm deep into the continent.<br />
The thickness of the Quaternary sediments is 40-50 m in<br />
the area under study. They are represented mostly by sandy<br />
and clayey soils in the first terrace and laida, respectively.<br />
The Varandey Peninsula is located in the zones of continuous<br />
and discontinuous permafrost with a total thickness d<br />
about <strong>25</strong>0 m. The average annual temperature of the perennially<br />
frozen ground (PFG) varies from 0 to 4 'C (Geocryology<br />
of the USSR 1988).<br />
On the first sea terrace (>95% of the area), PFG is continuous.<br />
Part-through taliks (zones of unfrozen ground) with a<br />
thickness of 5-10 m are confined to small lakes and rivers.<br />
Through taliks are located below large lakes and rivers. In the<br />
first sea terrace, PFG has a thickness of 40 to I00 m and is<br />
underlain by cooled salinized ground with lenses of saline<br />
water (cryopegs). Cryopeg salinity reaches 40-50 gll. The<br />
average annual temperature of the ground is the lowest (from<br />
-2 to -4 'C) at hillocky peatbogs and in areas without vegetation.<br />
The temperatures are the highest (from -0.5 to -2 'C)<br />
within wide swampy runoff bands, khasyreyas (thermokarst<br />
depressions), and the valleys of small streams.<br />
On the coastal part of the area (laida) PFG is discontinuous.<br />
The upper section (to a depth of 34 m) is composed d<br />
cooled ground at a positive temperature. Cooled salinized<br />
ground at negative temperatures (from -1.5 to -3 OC) composes<br />
the lower section. The freezing point of silty clay at a<br />
salinity of 2% is -3 OC. Therefore, frozen ground occupies<br />
only 50-90% of the laida area in spite of a continuous distribution<br />
of ground temperature from 0 to 4 'CC. The physical<br />
properties of a cooled ground correspond to those a thawed of<br />
ground. Therefore, we canstatethat PFG is discontinuous<br />
within the laida. The surface of the frozen ground is also bro<br />
ken by part-through and through taliks below saline rivers and<br />
lakes. The thickness of part-through taliks can exceed 50 m.<br />
Physical geological processes and formations, including<br />
cryogenic heaving, thermokarst and thermal erosion and p<br />
lygonal microrelief are widespread within the first sea terrace.<br />
Coastal erosion is observed at the high cliffs of the first sea<br />
terrace.<br />
METHODOLOGY<br />
Engineering geocryological zoning is the main method used<br />
in this study. It is performed according to the genetic principle<br />
and is based on landscape zoning. Topography, vegetation,<br />
and surface drainage are described for each landscape. The<br />
landscape zoning has been performed using data from field<br />
studies and the deciphering of aerial photographs. Geocryological<br />
areas were distinguished as a result of this zoning.<br />
Genesis and lithology of subsurface sediments, temperature,<br />
thickness and the occurrence of frozen, cooled and thawed<br />
ground and cryogenic processes, were determined for each<br />
area.<br />
The performed studies included the drilling of <strong>25</strong> boreholes,<br />
borehole thermometry, geophysical works (VES and<br />
electric profiling) and the analysis of ground composition and<br />
properties to a depth of I5 m.<br />
RESULTS AND DISCUSSION<br />
The map of engineering geocryological zoning of the<br />
Varandey Peninsula coast was constructed based on the<br />
performed studies. Figure I is only a fragment of this map.<br />
The zoning shows the relationship between the geomorphological<br />
level, genesis, lithology, landscape and average<br />
ground temperature.<br />
Three typical ground sections were distinguished de<br />
pending on the salinity: (I) the entire section (to a depth of 15<br />
m) is composed of salinized ground; (11) non-salinized and<br />
salinized ground occur in the upper (to a depth of 34 m) and<br />
lower sections, respectively; (Ill) non-salinized and salinized
sections, respectively. During the zoning it has been estab the upper section. As a result, although all grounds have<br />
lishedthatgeocryologicalconditionsdependdirectlyonland-negativetemperaturestheyexhibitthepropertiesofthawed<br />
scape conditions. Genetic types, lithology, temperature of the soils. Slightly salinized ground also occurs in the bottom secground<br />
and landscape type are shown on the map for each tions of the first sea terrace.<br />
distinguished area, the shading of each area corresponds b<br />
the average annual temperature of the ground.<br />
The section types with a different distribution of salinity ACKNOWLEDGMENTS<br />
(see above and Fig. 1) are shown by independent indices (I,<br />
II, Ill). This work supported was by the Russian<br />
Program and INTAS grant 2332.<br />
CONCLUSION<br />
REFERENCES<br />
The area under study includes continuous and discontinuous<br />
permafrost. Thawed ground and part-through taliks are local. Geocryology of the USSR. European Part. 1988. Moscow: Nedra (in<br />
The presence and non-uniform distribution of salinized cooled Russian).<br />
Rivkin F., Popova A., and Eospa A. <strong>20</strong>01. The Permafrost Condiground<br />
are responsible for specific geoc~ological<br />
tions along the Pechora Sea Coast (the Varandey Peninsula,<br />
in this area. This is especially evident within laidas, where<br />
European North of Russia), In <strong>Proceedings</strong> of Ist European<br />
the frozen ground is discontinuous because of high salinity in Permafrost Conference, Rome: 38.
8th International Canference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
An active rock glacier in the southernmost permafrost environment of the<br />
Alps (Argentera Massif, Italy): four years of surface ground temperature<br />
monitoring<br />
A. Ribolini<br />
Dipartimento di Scienze della Terra, University of Pisa, Via S. Maria 53, 56126 Pisa, Italy; ribolini@dst.unipi.it<br />
The Argentera Massif corresponds to the south-ernmost<br />
permafrost environment of the European Alps, being<br />
only 50 km far from the Mediterranean Sea (Fabre and<br />
Ribolini <strong>20</strong>03). This alpine sector is highly sensitive to<br />
environmental changes as demonstrated by last century<br />
glaciers disappearing. The surface ground temperature<br />
(SGT) is used in the analysis of the relationship between<br />
the near surface soil temperature and the nonconductive<br />
seasonal soil heat transfer processes, forcing<br />
permafrost growthldegradation (Kane et al. <strong>20</strong>01).<br />
Four years-ground thermal regime at the snow-ground<br />
interface of the Portette rock glacier will be analysed by<br />
means of hourly monitoring of soil temperature using a<br />
miniature digital data logger (DTL). In this rock glacier,<br />
located in the south-eastern sector of the Argentera<br />
Massif, previous geoelectrical sounding indicated the<br />
presence of permafrost between 5-<strong>20</strong> m depth (Fabre<br />
and Ribolini, <strong>20</strong>03).<br />
SGT data cover the November 1998-November <strong>20</strong>02<br />
period (Fig. I). Four onsets of autumn freezing and<br />
spring-summer thawing periods are recognizable. During<br />
winter months the temperature remains well below<br />
O”C, except for the <strong>20</strong>00-<strong>20</strong>01 winter, when it did not<br />
descend below -2°C. The October-May SGT reached<br />
the lowest value during 1998, when the highest annual<br />
“cold content” (degree day of frost, DDF) was experienced.<br />
The winter records show both a regular temperature<br />
decreaselincrease (1999-<strong>20</strong>00, <strong>20</strong>00-<strong>20</strong>01)<br />
and a .temperature trend characterized by several<br />
“spike-like” anomalies (I 998-1999,<strong>20</strong>01-<strong>20</strong>02).<br />
To better understand the ground thermal behavior,<br />
the hourly temperature change DT (“Clh) at the DTL<br />
depth (<strong>25</strong> cm) was calculated (Fig. 2). Four seasonal<br />
thermal regimes can be identified: summer thawing<br />
(ST), zero curtain (ZC), autumn-winter freezing (FR) and<br />
snow melt (SM).<br />
15 1999-00 <strong>20</strong>00-01 <strong>20</strong>01 -a2<br />
10<br />
5<br />
0<br />
-5<br />
-i 0<br />
-1 5<br />
L’<br />
Snow cover<br />
perlod<br />
-<br />
Snow cover<br />
period<br />
Figure 1, Daily air (grey) and Surface Ground Temperature<br />
The large hourly temperature change characterizing “Zero Curtain Effect” (ZC), i.e. the compensation of the<br />
the ST regime can be considered to represent the rapid upward heat loss associated with the drop in air temperaresponse<br />
of the active layer to weather fluctuations. tures by the freezing latent heat of the soil. The hourly<br />
The ground freezing thermal behavior is characterized temperature change of the autumn-winter FR regime<br />
by a lapse of time during which the temperatures show a shows positive (warming) and negative (cooling)“spike-<br />
reduced variation and maintain themselves at a constant like” fluctuations, that are sharp in the first and last winter,<br />
value below 0 “C. This behavior can be attributed to the suggesting atmosphere-ground interaction. At the end of
Ribolini, A,: Active rock glacier in the southernmost permafrost environment., .<br />
AT<br />
('cclhr)<br />
.b :":*<br />
ST<br />
ST<br />
/<br />
SGT<br />
no data<br />
. ..<br />
...<br />
. .<br />
I .<br />
, * '<br />
ZlSep OlJan 071ar 19Jul 08.Nov Web 24Yay 234111<br />
Figure 2. Hourly temperature change (DT, "Clhr) at DTL depth<br />
every FR regime a rapid increase in SGT occurs, followed<br />
by a close 0 "C temperature plateaux (SM). These sharp<br />
temperature rises are decoupled from the daily air temperature<br />
variations and could be referred to nonconductive<br />
processes such as snow melt-water infiltration.<br />
The 0 "C flat trace of several days, that follows every<br />
rapid increase (Fig.l), could represent the ice-water phase<br />
change. In order to evaluate the relations between the<br />
presence of snow on the rock glacier surface and non<br />
conductive processes, the ground-air thermal signal couplingldecoupling<br />
could be considered. The magnitude of<br />
ZC effect strongly depends on the soil water content, being<br />
longer in response to enhanced moisture. The relation<br />
between snow cover and the limitedlabsence of ZC for the<br />
1998 and <strong>20</strong>00 autumn suggests that the air temperature<br />
lowering affected a well drained ground without a consistent<br />
water content. During the 1999 and <strong>20</strong>01 FR regimes,<br />
the snow cover presence assured a relative water content<br />
inside the block voids, triggering and prolonging the ZC<br />
effect. The 1998-99 and <strong>20</strong>00-01 winter recordings (FR)<br />
clearly show coupled air-ground thermal traces, with<br />
trends evidently disturbed by abrupt SGT. Funnels in the<br />
thin snow cover likely allow the replacement of warm air<br />
present in the block voids by dense cold air descending<br />
from the atmosphere. Spike-like thermal anomalies are<br />
absent or very limited during 1999-00 and <strong>20</strong>00-01, because<br />
the snow cover isolates the ground from external air<br />
fluctuations. The warming magnitude relative to snowmelt<br />
water infiltration in the frozen soil (SM) mainly results from<br />
the infiltrated water volumeltemperature and soil porosity.<br />
The coarse-grained ground surface of the Portette rock<br />
glacier allows rapid infiltration of snow meltwater. To this<br />
non-conductive process corresponds a high thermal impact<br />
resulting in an up to 4-5 "C increase within a few<br />
days. Finally, the long-lasting snow cover until late spring<br />
insulates the ground from the incoming solar radiation and<br />
prevents the ground from heating up.<br />
The Portette rock glacier shows SGT trends characterized<br />
by non-conductive and seasonally driven heat<br />
transfer processes, i.e. release of latent heat during phase<br />
change, snow melt infiltration and air circulation. The<br />
ground-air temperature comparison suggests that the<br />
permafrost growthldegradation strongly depends on the<br />
timing of snow appearanceldisappearance at the rock glacier<br />
surface. Delayed and thin snow cover in autumn represents<br />
a forcing factor for permafrost maintenancelgrowing<br />
because it allows i ) strong penetration of<br />
cold into the ground, ii) lackinglreducing of the ZC effect<br />
and ii) downward funnelling of cold air.<br />
REFERENCES<br />
Fabre, 0. and Ribolini, A. <strong>20</strong>03. Limite actuelle des glaciers ro-chers<br />
actifs dans le massif de I'Argentera (Alpes Maritimes italiennes).<br />
Reunion du Societe Hydrotechnique de France, March <strong>20</strong>03,<br />
Grenoble.<br />
Kane, D.L., Hinkel, K. M., Goering, D.J., Hinzman L.D. and Outcalt,<br />
S.I. <strong>20</strong>01. Non-conductive heat transfer associated with frozen<br />
soil. Global and Planetary Change 29: 275-292.<br />
134
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Undercooled scree slopes covered with stunted<br />
abiotic factors to characterize the<br />
dwarf trees in <strong>Switzerland</strong> -<br />
phenomenon<br />
A. Risf, M. Phillips and K. Auerswald*<br />
Swiss Federal Institute for Snow and Avalanche <strong>Research</strong> (SLF), Davos Doe <strong>Switzerland</strong><br />
*Chair of Grassland Sciences, Department of Plant Sciences, Technical University of Munich- Weihenstephan, Freising,<br />
Germany<br />
INTRODUCTION<br />
Large-scale permafrost distribution can be estimated using roundings and negative if azonal permafrost is to occur. The<br />
numerical models simulating the spatial distribution of penna- energy balance at a site is determined by four energy balfrost<br />
in complex topography (Hoelzle et al. <strong>20</strong>01). The phe ance factors: radiation balance, ground heat flux, sensitive<br />
nornenon can exist sporadically far below the alpine timber- heat flux and latent energy flux.<br />
line as azonal permafrost (Haeberli 1978) at special locations<br />
with extreme shadow and a significant lower mean annual<br />
ground temperature than their surroundings. Sites satisfying<br />
this definition are called undercooled scree slopes by<br />
Wakonigg (1996) including locations which are not frozen<br />
perennially,butsignificantlylongerthantheirsurroundings.<br />
Undercooled scree slopes are characterized physiognomically<br />
by a coverage of stunted dwarf trees (Fig. I), whose<br />
growth is limited by the cold (Sieghardt et <strong>20</strong>00). al.<br />
Already in the 19th century an air stream through the<br />
voids of the coarse blocky material of the scree slope<br />
changing its direction seasonally was regarded as the causal<br />
process for the exceptionally cold temperatures. This can be<br />
explained by the relatively constant temperature of the air in<br />
the underground compared to the seasonal temperature fluctuations<br />
in the above-ground atmosphere: in summer the relatively<br />
cold air inside the scree slope sinks downwards<br />
through the big voids between the blocks and thereby draws<br />
warm atmospheric air through the upper openings of the scree<br />
slope. On its way through the underground the air cools<br />
down on contact with the cold blocks and comes out as an<br />
icy wind at the slope foot. In winter, when the rock is relatively<br />
warm, the air stream changes its direction: cold atrnospheric<br />
air streams into the lower openings of the scree slope,<br />
warms on the blocks, which thereby cool down, and ascends<br />
due to its lower density in order to escape into the atmosphere<br />
at the upper end of the scree slope. This theory<br />
only explains why the ground temperatures of an underigu~~<br />
1. ~ n d ~ ~ s~ree ~ ~ slop ~ l~r~~~ e d du Van in the ~wis~ ~~r~<br />
cooled scree slope are significantly lower than the surroundrnou~~a~n~.<br />
The stunte~ ~ w t~ee~ a ~ (white f r ~ i~~i€ate m ~ ~ ~ u~de<br />
ing air temperature in summer.<br />
€QQi~~ ~ ~ n d i ~ i ~ n ~ .<br />
However, out of this it can not be derived that the mean<br />
annual ground temperature of this site is lower than in the surroundings,<br />
because the lack of energy in winter might be<br />
compensated by the surplus of energy in summer. The presented<br />
theory can only be kept if the winter effect preponderates<br />
the summer effect. Therefore an asymmetry in the sys-<br />
tem has to be looked for. However, the energy balance of an<br />
undercooled scree slope has to be lower than in its sur-<br />
Our objective was to determine the abiotic factors of<br />
dercooled scree slopes. In <strong>Switzerland</strong> 36 sites were investigated.<br />
On the landscape scale, climatic, relief and ground<br />
parameters were determined with GIs-techniques and com-<br />
pared with the equivalent values of 495 reference sites with<br />
zonal forest. Ten of the undercooled scree slopes were stud-<br />
135<br />
un
ied in more detail by comparing each with its reference site<br />
with zonal forest in the immediate surroundings in respect b<br />
climatic, relief, surface and ground parameters. For that purpose<br />
each study slope was divided vertically into a high<br />
(HZ), middle (MZ) and low zone (LZ) and horizontally into a<br />
transect of stunted dwarf trees and one of zonal forest at the<br />
slope foot (Fig. 2).<br />
Rist, A. et al.: Undercooled scree slopes covered with stunted dwarf trees..<br />
the lower energy level will be conserved by thermal insulation.<br />
In spring this can result in a delayed snowmelt, Le. the<br />
undercooled scree slope will become snowfree later than the<br />
surroundings. Due to the higher short wave albedof snow in<br />
comparison with soil and vegetation this will further decrease<br />
the annual areal radiation balance of the undercooled scree<br />
slope as compared to its surroundings.<br />
Most of the investigated parameters differ only slightly<br />
between an undercooled scree slope and its surroundings<br />
and therefore the differences of the energy balances are also<br />
marginal. Nevertheless, the physiognomical forms of forest d<br />
the compared sites are different. Thus undercooled scree<br />
slopes are labile systems and therefore represent a rare phe<br />
nomenon and an ecosystem worthy of protection.<br />
Figure 2. Zoning of an investigated slope horizontally in functional<br />
transects, vertically in a high (HZ), middle (MZ), low zone (LZ) and<br />
the rock face.<br />
The investigated undercooled scree slopes are located in<br />
areas with a mean annual precipitation between 1700 and<br />
<strong>20</strong>00mm and at a mean altitude of over1000m a.s.1. The Osterreich 37.<br />
presence of undercooled scree slopes is therefore restricted b<br />
Wakonigg, H. 1996. Unterkuhlte Schutthalden. Arbeiten aus dem<br />
lnstitut fur Geographie der Karl-Franzens-Universitat Graz Beimountain<br />
regions, but not to an exceptionally cold zonal cli-<br />
trage zur Permafrostforschung 33: <strong>20</strong>9-223.<br />
mate. Undercooled scree slopes are characterized by special<br />
local factors: the slope and the rock face above (present in<br />
90% of cases) have a northerly aspect, are shady, and the<br />
slope inclination is at least <strong>20</strong>". The rock face - if present - is<br />
at least 75m high and 40" steep. These relief parameters<br />
cause a low annual radiation balance of about 2.3 GJlm2 on<br />
average over the slope profile. In the transect of the stunted<br />
dwarf trees respectively the undercooled scree slope the<br />
surface area of open scree (about 40%) is more than twice<br />
as high as in the proximate transect with zonal forest. There<br />
fore the absorption of radiation will be less in the first transect,<br />
as the short wave albedo of rock is higher than that of soil<br />
and vegetation. Furthermore the air convection flux through<br />
the coarse blocky material, which is important for the low energy<br />
balance of an undercooled scree slope (Delaloye and<br />
Reynard <strong>20</strong>01), can be ascribed to the large surface area af<br />
open scree and to the wide pore diameters (increasing from 4<br />
cm on average at the head of the scree slope to IOcm at its<br />
foot). The soil depth of the undercooled scree slope (8 cm on<br />
average) is less by about one third than in the surrounding<br />
zonal forest. The outcome of this is a reduced thermal insulation<br />
of the undercooled scree slope. In combination with the<br />
stronger air convection this may lead to a faster cooling of the<br />
undercooled scree slope in comparison to the surrounding<br />
forest in autumn. When the snow cover reaches about I m,<br />
REFERENCES<br />
Delaloye, R. and Reynard, E. <strong>20</strong>01. Les eboulis geles du Creux du<br />
Van (Chaine du Jura, Suisse). Environnement Pkriglaciaires.<br />
Notes et Comptes-Rendus du Groupe Regionalisation du Periglaciaires<br />
8: l 18-129.<br />
Haeberli, W. 1978. Special aspects of high mountain permafrost<br />
methodology and zonation in the Alps. <strong>Proceedings</strong> of the Third<br />
International Conference on Permafrost, NRC Ottawa, 1, 379-<br />
384.<br />
Hoelzle, M. Mittaz, C. Etzelmuller, B. and Haeberli, W. <strong>20</strong>01. Surface<br />
energy fluxes and distribution models relating to permafrost in<br />
European Mountain Permafrost areas: an overview of current<br />
developments. Permafrost and Periglacial Processes 12: 53-68.<br />
Sieghardt, H., Punz, W. and Reimitz, R. <strong>20</strong>00. Vergleichendanatomische<br />
Studien an Pflanzen der Eppaner "Eislocher".<br />
Verhandlungen der Zoologisch-Botanischen Gesellschaft in<br />
136
Formation and evolution of the Shalbatana Valley System, Mars: A case<br />
study for the interaction between magmatism and permafrost<br />
J. A. P. Rodriguez, S. Sasaki, *R. 0. Kuzmin and ** R. Greeley<br />
Department of Earth and Planetary Sciences, University of Tokyo, Japan<br />
*Vernadsky Institute of Russian Academy of Sciences, Moscow, Russia<br />
**Department of Geological Sciences, Arizona State University, Arizona, U.S.A<br />
INTRODUCTION<br />
A large concentration of seemingly collapsed features and<br />
associated oufflow channels are found between western Lunae<br />
Planum and western Arabia Term in the cratered highlands<br />
(units Npll and Np12) and the volcanic plains (unit Hr).<br />
Comparisons with terrestrial flood channels, such as those d<br />
Channeled Scablands in eastem Washington, suggest that<br />
the Martian channels were eroded by catastrophic floods.<br />
Most of the Martian channels originate from chaotic terrains<br />
interpreted to represent areas where the ground collapsed as<br />
water (confined under the permafrost) was released under<br />
high hydrostatic pressures within artesian basins. Understanding<br />
the geologic history of the Shalbatana Valley System<br />
(SVS) can improve our knowledge of groundwater recharging<br />
and extraction mechanisms, and the derivation d<br />
the history of valley formation in the highlands. Shalbatana<br />
Vallis was interpreted as an oufflow channel excavated by<br />
water flow from an ice-covered paleolake in Ganges<br />
Chasma (Nedell et al. 1987) and from catastrophically releases<br />
from confined aquifers as the permafrost seal (Clifford,<br />
1987) was disrupted at three locations, forming chaotic regions<br />
(Cabrol et al. 1997). Rodriguez et al. (<strong>20</strong>03) proposed<br />
that the excavation of SVS also involved water released<br />
from an extensive underground cavern system produced by<br />
magma and permafrost interaction. They also proposed an<br />
alternative hypothesis for the origin for the Shalbatana up<br />
stream chaotic region; they suggest that a highly degraded<br />
late Hesperian impact crater (SE part) collapsed over the<br />
putative cavernous system. In this work, we have used 128<br />
pixelsldegree MOLA DEMs to derive alternative hypotheses<br />
for the origin the Basin-A and its chaotic material, as well as<br />
for the origin of the flanking Valley system Ill (Fig. IA).<br />
THE BASIN A<br />
The abrupt widening of the main Shalbatana Valley occurs<br />
at the junction between VS-I and B-A (Fig. 1 B). This is<br />
also the transition of the main valley from incision into the<br />
Noachian plateau (bounded by unit Np12 to the east) and b<br />
the West by VSlll (Hesperian highland unit Hr). The chaotic<br />
terrain in 6-A was used as evidence that this basin resulted<br />
from collapse over an aquifer (Cabrol et al. 1997). The east<br />
margin of the main SVS valley shows a change in orientation<br />
where it bounds B-A, which resemble intercepted craters.<br />
The topographic region TLI partly bounds the basin at an ele<br />
vation of about 700 meters above TU and TL3 (Fig. 16). A<br />
channel (Chn) extends from VSI into B-A, the margins of<br />
which are partly bounded by TLI-A and TLI-B. This suggests<br />
that the topography was continuous prior to channel<br />
formation. The topographic and roughness characteristics d<br />
these terrains resemble those of Crater A (Fig. I6 and Fig.<br />
2). Moreover, unit TLI-C occurs along the flanks of the arcuate<br />
cliff section. These observations are consistent with this<br />
unit representing the floor of collapsed craters intercepted by<br />
SVS. The karst-like features (seen on the top edges of B-A),<br />
are consistent with collapse processes for this part of<br />
the<br />
valley and suggest that the permafrost might have been relatively<br />
shallow in the past andlor there are efficient mechanisms<br />
of vertical transport of volatiles from a deeper permafrost<br />
region, such as fractures produced by surface extension<br />
over intrusive nonemergent dikes. Structural control (Cabrol<br />
et al. 1997) over the main valley determinedthedeepand<br />
narrowU-shaped VSI, which is incised into the Noachian<br />
highlands. Structural control along the eastern flank of the<br />
main valley is suggested by the orientation, depth and slope<br />
characteristics and we propose that the abrupt widening of the<br />
main valley resulted from a change in lithologies. This hypothesis<br />
is consistent with the distribution of the chaotic terrain,which<br />
is rougher, more abundant, and topographically<br />
higher on the flanks of VSIII. This suggests that the chaotic<br />
material was shed from the VSlll upland. The position of the<br />
deepest and smoothest topographic unit (TL3) is consistent<br />
with its being a depositional area for material carried by the<br />
channel (Chn). We propose that flow from VS-I intercepted a<br />
series of collapsed craters in this region and possibly formed<br />
a crater-lake system, which might have extended as far as<br />
the end of VS-II. The geological materials composing the<br />
western margin of this lake system were apparently volatile<br />
rich and suggest that the pemafrost at a regional scale was/<br />
137<br />
is unequally distributed andlor has undergone different<br />
logical processes.<br />
gee
THE VSlll COMPLEX<br />
Rodriguez, J.A.P. el A+: Formation and evolution of Elre Shalbatana Valley System ...<br />
It has been proposed (Cabrol et al. 1997) that water from<br />
the B-A region successively drained and excavated the dl<br />
and d2 channels in the Hesperian plateau (Fig. 2). These two<br />
channels were abandoned following the deepening of the<br />
main channel towards the Simud valleys (Cabrol et ai.<br />
1997). We have observed that this region forms a complex<br />
system of valleys, which includes rimless quasi-circular de<br />
pressions, elongate depressions, pit chains and deep U-<br />
shaped valleys. The roughness of this region reveals a high<br />
frequency, low amplitude component, which would be consistent<br />
with extensive thermokarstic activity and other sub<br />
surface degradation processes (Rodriguez et al. <strong>20</strong>03; Kuzmin<br />
et al. <strong>20</strong>02).<br />
The d2 system: The contact between the d2 system and<br />
B-A forms a wide valley (Fig. 2, yellow dotted line), which<br />
narrows and is incised by a relatively narrow, southdipping<br />
central region (d2a). Channel d2a joins with d2b to the north<br />
forming a deep U-shaped valley. B-A drainage by the dl and<br />
d2 channel systems is consistent with the temporal ponding<br />
of VS-1 , B-A andVS-11, which was subsequently lowered<br />
as flow toward the Simud valley occurred (Kuzmin et al.<br />
<strong>20</strong>02).<br />
We propose that in a first stage flow took place from B-A<br />
to the north, forming a drainage channel of which only the<br />
northem d2b section remains. After drainage from B-A<br />
ceased the northern margin of B-A became unstable and recession<br />
occurred preferentially headward along the preexisting<br />
drainage channel. Recession happened mainly as debris<br />
flows, fed by material shed from the valley flanks to its central<br />
region, flowed towards B-A, which resulted in valley<br />
widening, excavation of the central channel d2a, and the<br />
deposition of most of the chaotic unit TL2.<br />
The dl system: This system forms an N-trending series<br />
of depressions, which includes wide shallow regions (Fig. 2,<br />
WSR), narrow enclosed-and elongate-regions (Fig. 2,<br />
NEER), and deep, rimless quasicircular depressions (Fig. 2,<br />
RQCD). This association does not seem to be the result d<br />
surface flow and it is more consistent with subsidence related<br />
to long-lasted subsurface degradation, possibly by water<br />
flowing through underground conduits (Rodriguez et al. <strong>20</strong>03).<br />
Other features in the region consistent with large subsurface<br />
processes include the floor of Crater A, which shows surface<br />
modifications and an elongate depression that extends south<br />
from the margin of crater A, forming a sequence of pits, which<br />
shallow up and become narrower to finally disappear near the<br />
margin of the VS 11. The presence of dike like ridges at the<br />
bottom of grabens located within the VSlll complex are suggestive<br />
of the observed featured to be genetically related ts<br />
interactions between intrusive magmatism and permafrost<br />
(Rodriguez et ai. <strong>20</strong>03).<br />
FINAL REMARKS<br />
These regions of the SVS demonstrate the importance d<br />
lithological and structural controls over the morphological<br />
characteristics of highland valleys. Our observations suggest<br />
that processes possibly related to debris flows and hermokarstic<br />
degradation produced the knobby material in the<br />
chaotic region within B-A. The observing evidence for extensive<br />
underground denudation apparently had resulted in different<br />
degrees of subsidence, which might have been related b<br />
thedevelopment of cavernous systems. These issues are<br />
important to understand the complexity of hydrological processes<br />
involved in the formation of the SVS, to quantify the<br />
amount of water involved, and the amount of material eroded<br />
during its excavation and to stress the complexity and great<br />
age range of the processes involved in the Martian perma-<br />
REFERENCES<br />
Nedell, S.S., Squyres, S.W. and Andersen, D.W. 1987. Origin and<br />
evolution of the lavered deDoslts in the Valles Marineris. Mars.<br />
lcarus 70: 409-441.<br />
Clifford, S.M. 1993. A model for the hydrologic and climatic behavior<br />
of water on Mars. Geoahvs. Res. Lett.. Vol. 98 No.E6. 19073-<br />
I ,<br />
11016.<br />
Cabrol, N.A., Ednmon, A. G. and Dawidowicz, G. 1997. Model of outflow<br />
generation by hydrothermal underpressure drainage in<br />
volcano-tectonic environment, Shalbatana Vallis (Mars), Icarus,<br />
1<strong>25</strong>: 455-464.<br />
Rodriguez, J.A.P., Sasaki, S. and Mi amoto, H. <strong>20</strong>03. Nature and<br />
Hydrological relevance of the Shaybatana complex underground<br />
cavernous s stem ,Geophys. Res. Lett., Vol. 30 No.6, 1304, 10.<br />
Kuzmin, R.O., ireeley, R. and Nelson, D. M. <strong>20</strong>02. Mars: The morhological<br />
evidences of late Amazonian water activity in Shalgatana<br />
Vallls. Lunar and Planetary Science conference XXXIII.<br />
Abstract 1087.<br />
138
8th International Conference on Permafrost Extended Abstracts an Current <strong>Research</strong> and Newly Available ~~~~~~~~~~~<br />
Distribution of ramparted ground-ice depressions, Llanpumsaint, southwest<br />
Wales.<br />
N. Ross, C. Harris, P. Brabham and *S. Campbell<br />
School of Earth, Ocean and Planetary Sciences, Cardiff University, Cardiff, Wales<br />
*Department of Earth Sciences, Countryside<br />
Council for Wales, Bangor, Wales<br />
INTRODUCTION<br />
Geomorphological features recognised as ‘ramparted groundice<br />
depressions’ are well represented in the Welsh landscape.<br />
These features have been interpreted as some of the<br />
finest examples of relict open-system pingos in the UK,<br />
formedwithindiscontinuouspermafrostduringtheLate De<br />
vensian or Loch Lomond Stadia1 (GS-1; Watson 1972).<br />
Landforms include horseshoe-shaped, circular or complex<br />
ramparts, enclosing water- andlor peat-filled basins.<br />
Although they are distributed throughout Wales (Fig. l),<br />
previous research indicates a remarkable density in the<br />
southwest, particularly within the tributaries of the Afon Tefi<br />
(e.g. Cledlyn Valley), northwards towards Aberystwyth<br />
(Watson 1972).<br />
Figure 1: Distribution of ramparted ground-ice depressions<br />
Wales (Adapted from Ballantyne and Harris 1994).<br />
in<br />
The known distribution of ramparted depressions in Wales<br />
is however, thought to represent only a proportion of the total<br />
number of potential sites. <strong>Research</strong> aimed at establishing their<br />
distribution, morphology, structure and origin has been initiated<br />
to develop a framework for their conservation. This paper<br />
concentrates on the spatial distribution and density d<br />
these features, presenting preliminary results from a representative<br />
location where ramparted depressions are documented,<br />
but no rigorous investigation has previously been<br />
conducted.<br />
LOCATION AND RESULTS<br />
Ramparted depressions were identified near Llanpumsaint by<br />
Watson and Watson (1974) in aseries of papers on ramparted<br />
depressions in southwest Wales during the 1970s.<br />
Their research focused predominantly on the features of the<br />
Cledlyn and Cletwr valleys however, and although other locations<br />
were illustrated in maps in various publications, the<br />
exact map references of these sites were never published.<br />
Re-identification and geo-referencing of these “dots on maps”<br />
is therefore a current research priority.<br />
The ramparted depressions of Llanpumsaint are located<br />
within an area of approximately km2 <strong>20</strong> in the tributaries of the<br />
Afon Gwili at altitudes between 90-150 m. Aerial photography<br />
has enabled the identification of numerous impressive landforms<br />
with ramparts enclosing depressions and impounding<br />
marshland. In excess of I00 clearly defined ramparts are<br />
identifiable, not including areas of complex micro-topography<br />
that may also represent ramparted depressions. A particularly<br />
impressive subset of features is located to the south of Llanpumsaint<br />
at SN 4<strong>20</strong>275.<br />
Preliminary mapping of this locality indicates a remarkable<br />
density (45 perkm2)of identiable ramparts (Fig. 2),<br />
representing approximately <strong>25</strong>% of the total number of identifiable<br />
ramparted depressions proximal to<br />
139
Ross: N. et ai.: Distribution of ramparted ground-ice depressions ...<br />
Llanpumsaint. The relationship between the distribution d graphical setting of the Llanpumsaint ramparted depressions<br />
these features and water courses is of note, with well- with those of the Cledlyn valley means that a similar mechadeveloped<br />
forms located in the areas adjacent or to, between, nism of formation can be tentatively inferred.<br />
channels. Without the support of field verification however, the Tne assignment of all ramparted depressions recorded in<br />
impact of channel modification for agricultural purposes cannot Wales to the Loch Lomond Stadia1 (ES-I) period must be<br />
be assessed, and the current network of water courses may questioned. Ground-ice mounds producing ramparted depresnot<br />
reflect the natural state. Like the features of the Cledlyn sions of this number, size and density are unlikely to have<br />
valley, those at Llanpumsaint are clustered on valley bottoms formed within 1 ka.<br />
and on low gradient slopes. This common morphology and To test these initial hypotheses, more detailed investigageographical<br />
setting suggests a common mechanism of for- tions are planned to:<br />
mation.<br />
i. produce a complete inventory of ramparted depression<br />
localities in Wales using aerial photography;<br />
WIDER IMPLICATIONS AND FURTHER RESEARCH<br />
ii. define the extent of ramparted depressions within<br />
identified sites, and assess their relative conservation<br />
status;<br />
Preliminary fieldwork and a review of aerial photography reiii.<br />
investigate selected sites utilising a suite of geo.<br />
sources confirm the initial assessment that the distribution d<br />
physical techniques and vibrocoring to determine inramparted<br />
depressions in Wales is much more extensive and<br />
of a higher density than previously recorded. This has imternal<br />
structure and stratigraphy.<br />
portant implications for the genetic origin and age of these<br />
features.<br />
The context necessary for the formation of contemporary REFERENCES<br />
ground-ice mounds depends not only on climate, but also on<br />
Ballantyne, C.K. and Harris, C. 1994. The Periglaciation of Great<br />
an array of geological and hydro-geological factors. The re<br />
Britain. Cambridge University Press, Cambridge.<br />
markablenumberanddensityoframparteddepressions in Gurney, S.D. 1995. A reassessment of the relict Pleistocene “pinsouthwest<br />
Wales is therefore indicative of a combination d gos” of west Wales: hydraulic pingos or mineral palsas Quafactors<br />
that facilitated their development. Potentially, these in- ternary Newsletter 77: 6-16.<br />
clude the interaction of i) topography, ii) superficial geology, Watson, E. 1972. Pingos of Cardiganshire and the Latest Ice Limit.<br />
Nature 236: 343-344.<br />
iii) structural geology, iv) the extent and influence of the Late<br />
Watson, E. and Watson, S. 1974. Remains of pingos in the Cletwr<br />
Devensian ice sheet, v) subsurface hydrology, and vi) gm basin, south-west Wales. Geografiska Annaler 56A: 213-2<strong>25</strong>.<br />
thermal flux.<br />
Recently, a hypothesis that the ramparted depressions d<br />
the Cledlyn and Cletwr valleys may represent the relict<br />
forms of mineral palsas has been propagated in the literature<br />
(Gurney 1995). In our opinion the origin of these features remains<br />
an open question. Investigation of their internal structures<br />
remains critical in addressing this issue, together with<br />
the location, morphology, volume of the ramparts, hydrological<br />
setting and distribution of sediments susceptible to ice segregation.<br />
The apparent similarities of morphology and g e<br />
140
Groundwater under deep permafrost conditions<br />
T. Ruskeeniemi, LAhonen, M. Paananen, R. Blomqvisf, P. Degnan, ** S. K. Frape, *** Mensen, **** K.Lehto, 1. Wiksfrom,<br />
***** L. Moren, 1. Puigdomenech and ****** M. Snellman<br />
Geological Survey of Finland, Espoo, Finland; * UK Nirex Ltd. , UK; ** University of waterloo, Waterloo, Canada<br />
*** Ontario Power Generation, Toronto, Canada; **** Posiva Oy, Olkiluoto, Finland<br />
***** Svensk Karnbranslehantering AB (SKB), Stocholm, Sweden<br />
****** Consulting Engineers Saanio & Riekkola, Helsinki, Finland<br />
INTRODUCTION<br />
According to the geologic disposal concept in Scandinavia,<br />
spent nuclear fuel will be sealed in corrosion-resistant<br />
and watertight canisters, and then emplaced in holes<br />
drilled at the bottom of tunnels excavated into the bedrock.<br />
The canisters are to be surrounded with bentonite clay,<br />
which expands when it absorbs water. In Scandinavia the<br />
depth of any repository is to be approx 500 m below<br />
ground surface. When the last canisters have been emplaced,<br />
the tunnels are to be filled, e.g., with a mixture of<br />
bentonite and crushed stone, and the shafts leading to the<br />
repository closed.<br />
In assessing the safety of the geologic disposal concept,<br />
it is essential to assess what physical and chemical<br />
conditions could evolve within the repository and surrounding<br />
environment. The objective of performance and<br />
safety assessments is to demonstrate the long-term safety<br />
of a repository for long-lived radioactive waste and most<br />
management agencies consider scenarios up to I my into<br />
the future.<br />
During at least the past 2 million years, repeated cycles<br />
of glaciation have occurred in the Northern Hemisphere.<br />
At present, a major part of the area to the north of<br />
latitude 60"N is under permafrost Conditions, although<br />
within Fennoscandia at the same latitudes, temperate<br />
conditions prevail. Climate models based on astronomical<br />
forcing predict a cold period after approximately 60 000<br />
years. Cold, dry periods favour the formation of periglacial<br />
conditions and the development of deep permafrost, such<br />
that permafrost is expected to reoccur in northern and<br />
central Europe and North America.<br />
In order to predict the long-term performance of a geologic<br />
repository the influence of periglacial and glacial<br />
conditions must be considered. A key goal of this research<br />
project is to enhance the scientific basis for safety and<br />
performance assessment and to derive constrained permafrost<br />
scenarios relevant to the evolution of crystalline<br />
groundwater flow systems.<br />
A number of issues were identified as possible targets<br />
of study: 1) Freezing rate and depth extent of permafrost<br />
in crystalline bedrock; 2) Cold saline water segregations<br />
(cryopegs) and their role in transport processes, 3) Role of<br />
non-frozen areas (taliks) as pathways for groundwater<br />
flow, 4) Aggregation of solid methane hydrates (clathrates),<br />
and 5) Effects of freezing on bedrock stability.<br />
All these aspects are related to the long-term stability<br />
of hydrogeological and hydrogeochemical conditions.<br />
Some of these phenomena are reviewed at length in the<br />
literature and much is known about their role in shallow<br />
systems or in sedimentary environments. However, much<br />
less has been published regarding cryogenic processes<br />
and hydrogeology in crystalline rock under deep permafrost<br />
conditions.<br />
This PERMAFROST project is a co-operative international<br />
undertaking jointly funded by the participants from<br />
Finland (Geological Survey of Finland and Posiva), Sweden<br />
(Svensk Karnbranslehantering, SKB), Great Britain<br />
(UK Nirex Ltd) and Canada (Ontario Power Generation<br />
and University of Waterloo). After a short feasibility phase,<br />
the research at Lupin started in late <strong>20</strong>01 (Ruskeeniemi et<br />
al. <strong>20</strong>02).<br />
THE STUDY SITE<br />
Although the occurrence of permafrost is widespread in<br />
the Northern Hemisphere today, there are only a few locations<br />
at which permafrost in crystalline terrain can be investigated.<br />
Such is the case at the Lupin gold mine in Canadian<br />
Arctic (Nunavut Territory), approximately 1300 km<br />
north of Edmonton, and some 80 km south of the Arctic<br />
Circle. The distance to the White Sea in the north is about<br />
300 km.<br />
The site is within the continuous permafrost zone. Today<br />
the depth of the permafrost in the area extends to<br />
depths between about 400 and 600 m. The temperature<br />
measurements conducted in the mine indicate that permafrost<br />
persists down to 541 m. It is assumed that much<br />
of the deep permafrost in this part of Canada has been<br />
formed during the Holocene, i.e., during the past 10,000<br />
years.<br />
The region is located in the continental sub-arctic climate<br />
zone well above the timberline. The vegetation is<br />
141
typical for tundra: grass and low-growing shrub species<br />
and dwarf birch. The mean annual air temperature is<br />
-1 1 "C, and the annual extreme values range from 49°C<br />
to +31 "C. The annual mean precipitation is low, about 270<br />
mm.<br />
The Lupin gold deposit is located in an Archean<br />
metaturbidite sequence. The formation of the ore is dated<br />
at 2.5 Ga. The major lithological units at the site are crystalline<br />
rock types, phyllite and quartz-feldspar gneiss.<br />
However, their sedimentary counterparts mudstone and<br />
graywackelquartzite are locally preserved and recognizable.<br />
The gold ore is hosted by an amphibolitic iron formation.<br />
RESEARCH AT LUPIN<br />
To support an evaluation of the feasibility of the site and to<br />
develop a preliminary conceptual geosphere model, the<br />
research at Lupin began with the compilation of the available<br />
historical information on the lithostratigraphy, structural<br />
geology, hydrogeology, mine hydrology, Quaternary<br />
geology, etc. The mine site investigation focused on the<br />
collection of groundwater samples from the mine at existing<br />
boreholes and drift-exposed discontinuities.<br />
In general the mine is dry (mine discharge about 50 000<br />
m3/a), with the majority of the inflow occurring from a few<br />
long exploration boreholes and, to a lesser extent, from<br />
certain fracture zones. One of the fracture zones is repeatedly<br />
cut by the spiral mine access ramp, which provided<br />
an opportunity to collect dripping water to the 680 m<br />
level at which the seepage apparently drains. Water samples<br />
were obtained from a vertical cross section between<br />
the 87 m and 1 I30 m levels.<br />
As suspected, all the samples from frozen levels turned<br />
out to be contaminated by Na-CI brines used for drilling in<br />
permafrost. In comparison, the boreholes at lower levels<br />
provided insight into the chemistry of the deep groundwaters<br />
below the permafrost. These waters are brackish to<br />
saline, Na-CI dominant, with dissolved solids (TDS) ranging<br />
between 2 to 29 glL and lacking all indication (high<br />
S04, NO3 and tritium) of contamination.<br />
One of the key activities at Lupin is to explore whether<br />
the outfreezing process can result in the formation of a saline<br />
groundwater layer at the base of the permafrost in<br />
crystalline bedrock. An electromagnetic sounding carried<br />
out in the area gave indications of a subhorizontal conductor<br />
in the depth range of 400 to 700 m. It is assumed<br />
that this conductor would mark the location of the base of<br />
the permafrost. According to model calculations, a saline<br />
water layer with TDS of 5-30glL could explain the observed<br />
anomaly. Parallel to the field research, laboratory<br />
freezing experiments were carried out at the University of<br />
Waterloo, Canada with various groundwaters collected<br />
from Canada and Scandinavia. In addition to detailed information<br />
on the behaviour of major and trace elements<br />
and isotopic fractionation during outfreezing, the experi-<br />
ments demonstrated that it was difficult to generate significant<br />
volumes of concentrated fluids from the small vol-<br />
Ruskeeniemi. T, et al.: Groundwater under deep permafrost,,,<br />
ume of groundwater available in crystalline bedrock. This<br />
result appears to contradict the results from the geophysical<br />
field study.<br />
To resolve this dilemma, and to support other hydrogeochemical<br />
studies, two 400 to 500 m long low-angle research<br />
boreholes were drilled from mine adits in late <strong>20</strong>02<br />
in an attempt to intercept fractures close to the base of the<br />
permafrost that might contain uncontaminated groundwater.<br />
The preliminary results suggest that the groundwater<br />
table has declined and unsaturated conditions may exist<br />
below the permafrost. Similar information is determined<br />
from hydraulic pressure head measurements in boreholes<br />
on the 890 and 1130 m levels. Currently it is too early to<br />
say whether this is typical for deep permafrost conditions<br />
in an area with limited recharge and low hydraulic gradients,<br />
or if it is simply the effect of <strong>20</strong> years of mine operation.<br />
A small amount of groundwater is seeping into the<br />
new boreholes from the overlying permafrost. Geochemical<br />
and isotopic geochemical charaterization of these<br />
samples is underway.<br />
CONCLUSIONS<br />
The overall goal of the studies at the Lupin mine is to<br />
study permafrost features and processes, which could<br />
have relevance in assessing the performance and safety<br />
of a repository at depth. Thus information obtained must<br />
be transferable from observations in the Lupin mine to a<br />
generic nuclear waste repository in a similar geological<br />
environment. The research at Lupin will continue to address<br />
the hydrogeochemical questions, but major efforts<br />
are also directed towards improving the understanding of<br />
the hydrogeology below the permafrost, in particular, the<br />
influence of taliks on groundwater flow conditions.<br />
142<br />
REFERENCES<br />
Ruskeeniemi, T., Paananen, M., Ahonen, L., Kaija, J., Kuivamaki,<br />
A,, Frape, S., Moren, L., and Degnan, P. <strong>20</strong>02. Permafrost at<br />
Lupin: Report of Phase I. Geological Survey of Finland, Nuclear<br />
Waste Disposal <strong>Research</strong>. Report YST-112,59 p.
A new approach to dating firn accumulation in an ice cave in the Swiss<br />
Jura mountains<br />
F. Schlatfer, M. Sfoffel M. Monbaron and *M. luetscher<br />
Department of Geosciences, Geography, University of Fribourg, Fribourg, <strong>Switzerland</strong><br />
*Swiss Institute for Speleology and Karstology (SISKA), La Chaux-de-Fonds, <strong>Switzerland</strong><br />
INTRODUCTION<br />
Located at 1359 m a.s.1. on the south east side of the<br />
Swiss Jura range (BiereND, 6”17’50”/ 46”33’47”), a collapsed<br />
doline with a diameter of about 40 m acts as an<br />
important source of snow accumulation during the winter<br />
season. The particular morphology and the sufficient<br />
depth (-45m) of the cavity enable the trapping of cold air.<br />
This causes the freezing of infiltration water and the<br />
transformation of the snow into ice. Over sixty ice caves<br />
are recognized in the Swiss Jura. The St Livres ice cave is<br />
the second largest in the range. In 1976, its volume was<br />
estimated at 3,500 m3 (Gigon 1976). Due to the significant<br />
melting of ice in the recent past, its volume is now reduced<br />
to approximately one-third (Le. 1<strong>20</strong>0 m3).<br />
Next to its vertical pit, the cavity has a horizontal development<br />
of 60 meters (Fig. l), ice flows along for about<br />
40 m, leading to the lower part of the cavity. In this place,<br />
an overhanging ice wall forms the front of this small glacier.<br />
Thus it is possible to observe an almost continuous<br />
outcrop of the glacier. The presence of organic matter in<br />
the ice allows the identification of different layers. Moreover,<br />
great number of branches and trunks are trapped in<br />
the ice (Fig. 2). In several places, deformations of the ice<br />
layers confirm the presence of significant flow of the ice<br />
filling. This flux is also responsible for the transport of<br />
woody material that fell down in the accumulation zone,<br />
that is, tree trunks and branches slowly movingto the<br />
deepest part of the cavity.<br />
In the past, the front of the glacier was completely covered<br />
by a big stalagmite of ice forming a dome. Freezing<br />
water infiltration originating from the roof of the cavity<br />
formed this ice. In the last ten years however, the ice cave<br />
melted considerably and the ice dome totally disappeared.<br />
Furthermore, the front of the glacier also retreated, liberating<br />
trunks and branches. The ice contains essentially<br />
four tree species, namely European beech (Fagus sylvatica<br />
L,), Great maple (Acer pseudoplatanus L.), European<br />
silver fir (Abies alba Mill.) and Norway spruce (Picea<br />
abies (L.) Karst.). In contrast to the wood trapped in the<br />
ice, only the two conifer species are abundant in the local<br />
forest next to the ice cave.<br />
OBJECTIVES<br />
This study aims to date (relative age) all ice layers with<br />
woody material. Dendrochronological methods (Schweingruber<br />
1996) are used to cross-date the single layers<br />
containing subfossil tree trunks and branches. Analysis of<br />
living trees outside the ice cave will reveal the link between<br />
the living and the subfossil samples, permitting an<br />
absolute dating of the single layers. This work is expected<br />
to provide important data on the process of firn accumulation<br />
in ice caves.<br />
METHODOLOGY<br />
First, the front of the glacier in the ice cave is mapped in<br />
detail. Special focus is placed on the identification of different<br />
accumulation layers and the position of some 60<br />
143
subfossil trunks and branches of Norway spruce (Picea<br />
abies (L.) Karst.) trapped in the ice. Cross sections of<br />
every tree are made using a chainsaw. In the laboratory,<br />
the samples are dried and polished using sandpaper. In<br />
order to determine the age and the yearly increment of the<br />
samples, tree-ring widths are measured, detrended and<br />
averaged (Schweingruber 1996). The resulting series of<br />
the 60 sections are then crossdated in order to build a<br />
floating chronology (relative age).<br />
In a further step, living trees outside the ice cave are<br />
sampled in order to obtain growth patterns of undisturbed<br />
living trees in the neighbouring forests. From each tree, at<br />
least two cores are extracted with increment borers. The<br />
resulting reference curve covers approximately the last<br />
<strong>20</strong>0 years.<br />
Finally, the absolute reference curve of living trees is<br />
cross-dated with the floating chronology of trapped wood<br />
inside the cave in order to close the gap between the<br />
youngest subfossil and the oldest living trees.<br />
Schlatter, F. et al.: A new approach to dating firn accumulation<br />
CONCLUSION<br />
Dendrochronological investigations in ice caves have only<br />
rarely been attempted to date ice layers and the accumulation<br />
of firn. Nevertheless, this approach will help to better<br />
understand the processes at the origin of such ice filling<br />
and will significantly improve our knowledge of these particular<br />
permafrost occurrences, without distur-bing the<br />
subterranean environment too much. Preliminary results<br />
will be presented on the poster.<br />
REFERENCES<br />
Audetat M. and Heiss G. <strong>20</strong>02. lnventaire speleologique du Jura vaudois,<br />
partie ouest. Acadernie suisse des sciences naturelles, La<br />
Chaux-de-Fonds: IRL Renens<br />
Gigon R. 1976. lnventaire speleologique du canton de Neuchitel. Societe<br />
helvktique des Sciences Naturelles. Neuchitel.<br />
Schweingruber, F.H. 1996. Tree Rings and Environment. Dendroecology.<br />
Swiss Federal Institute for Forest, Snow and Landscape<br />
<strong>Research</strong> WSLIFNP. Birmensdotf:<br />
144
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Rock glacier inventory of the Adamello Presanella massif (Central Alps,<br />
Italy)<br />
R. Seppi, *C. Baroni and **A. Carton<br />
Dipartimento di Scienze della Terra, Universifa di Pavia, Pavia, lfaly and Museo Tridenfino di Scienze Naturali, Trenfo, Italy<br />
*Dipartimento di Scienze della Terra, Universia di Pisa and CNR, lsfitufo di Geoscienze e Georisorse, Pisa, lfaly<br />
**Dipartimento di Scienze della Terra, Universifh di Pavia, Pavia, Italy<br />
The rock glacier inventory of the Adamello Presanella rock glaciers facing<br />
N, NE and NW is considerably lower<br />
Group (Central Alps, Italy) was carried out by means of than the elevation those facing S, SE and SW (Fig. 1). In<br />
field and topographic surveys as well as aerial photograph the activelinactive forms this difference is <strong>20</strong>2 m, whereas<br />
analysis. The study area is characterized by a series of in the relict ones it is 89 m.<br />
valleys descending from the summit of the massif with a<br />
centripetal pattern. Geomorphology is dominated by mass<br />
N<br />
wasting (talus slopes, debris cones, debris-flow cones and<br />
lobes), and glacial (erosional landforms and moraines)<br />
2700 m a.s.1.<br />
and periglacial (mainly rock glaciers) landforms. Lithology<br />
is characterized by the Adamello-Presanella-Monte Re di<br />
Castello intrusive rock complex (tonalites and granodiorites).<br />
The metamorphic basement and volcanic and<br />
sedimentary rocks of the Permian-Cretaceous sequence<br />
outcrop at the external margins of the batholite. The<br />
mountain summit area hosts the largest glacier of the Italian<br />
Alps (Adamello Glacier, 17.66 km2); more than 90 glaciers<br />
are to be found in the northern and central sectors of<br />
the massif.<br />
In the study area 216 rock glaciers have been identified<br />
and mapped by means of GIS (ArcView 3.2). They<br />
have been divided into (a) activelinactive (83 rock glaciers,<br />
38%) and (b) relict (133 rock glaciers, 62%), according<br />
to Barsch (1996). The average elevation of the<br />
front of activelinactive rock glaciers is 2493 m a.s.1. (mini-<br />
S<br />
mum value 2180 m a.s.1.). The average of the maximum<br />
Lower limit activelinactive<br />
elevations of the same rock glaciers is 2614 m a.s.1.<br />
rock glacters<br />
I I Lnwer limit relict<br />
(maximum value 2940 m a.s.1.). The average elevation of<br />
rock glaciers<br />
the front of the relict forms is <strong>20</strong>71 m a.s.1. (minimum<br />
value 1700 m a.s.1,). The average maximum elevations of Figure 1, Altitude of the fronts of activelinactive and relict rock glaciers<br />
the same rock glaciers is 2183 m a.s.1. (maximum value compared with aspect. The number of forms for each orientation is<br />
2650 m a.s.1.). Activelinactive rock glaciers are on average shown.<br />
291 m long and 163 m wide, whereas the relict ones are<br />
on average some 290 rn long and I99 m wide. The average<br />
area covered by rock glaciers is 0.041 km2, varying<br />
from 0.039 km2 for the activelinactive forms and 0.043 km2<br />
for the relict ones. The largest form (A75 in the inventory),<br />
located in Val di Lares, is a lobate type and extends over<br />
0.27 km2.<br />
In terms of orientation, 61% of rock glaciers face N, NE<br />
and NW, whereas 17% are exposed to the S, SEI SW.<br />
The remaining ones face E (13%) and W (9%), respectively.<br />
The elevation of the front of activelinactive and relict<br />
Measurements of the displacement of rock blocks from<br />
two active rock glaciers are in progress. Preliminary survey<br />
has been carried out on each of them by means of a<br />
theodolit. The position of <strong>25</strong> boulders marked with screw<br />
anchors was checked one year after installation on a rock<br />
glacier located in the upper Val di Genova (A37 in the in-<br />
ventory). This tongue-shaped rock glacier, which ranges in<br />
elevation from 2750 to 2880 m a.s.l., faces SW, is some<br />
280 m long, 100 m wide and covers an area of about<br />
0.023 km2. The movement of the single monitored boul-<br />
145
ders ranges between 0.05 m and 0.21 m. The boulders<br />
scattered on the frontal and left-sided portions of the rock<br />
glacier (points 1-15) are subject to greater displacements,<br />
whereas the higher sector is less dynamic (Fig. 2). In any<br />
case, the measured movements are rather slow (cf.<br />
Barsch 1996, Kaufmann 1996) and generally oriented<br />
along the slope's maximum dip. The surface velocity surveyed<br />
in the second rock glacier, which is located in Val<br />
d'Amola (A42 in the inventory), ranges between 0.02 and<br />
0.16 mlyear, with flow vectors oriented in a rather homogeneous<br />
way. The eastern portion of the front appears to<br />
be more dynamic (0.12-0.15 mlyear).<br />
Seppi, R. et 81,: Ruck glacier inventory of the Adamello Presanelia massif.<br />
riod. On the other hand, more forms (about 30) are placed<br />
immediately outside the boundaries attained by glaciers<br />
during the LIA and do not show any relationship with the<br />
glacial deposits of this period. It may therefore be inferred<br />
that among activelinactive rock glaciers the most recent<br />
forms are ascribable to a period subsequent to the LIA,<br />
whereas the other forms are attributable to coeval or preceding<br />
Holocene phases.<br />
REFERENCES<br />
Barsch, D. 1996. Rock glaciers: indicators for the present and former<br />
geoecology in high mountain environments. Berlin: Springer-<br />
Verlag.<br />
Frauenfelder, R., Haeberli, W., Hoelzle, M. and Maisch, M. <strong>20</strong>01. Using<br />
relict rock glaciers in GIS-based modelling to reconstruct<br />
Younger Dryas permafrost distribution patterns in the Err-Julier<br />
area, Swiss Alps. Norsk geog. Tidsskr. 55: 195-<strong>20</strong>2.<br />
Kaufmann, V. 1996. Geomorphometric monitoring of active rock glaciers<br />
in the Austrian Alps. High Mountain Remote Sensing Cartography:<br />
Proc. Intern. Symp., Karlstadt-Kiruna-Tromsra, 19-29<br />
August 1996: 97-113.<br />
Lambiel, C. and Reynard, E. <strong>20</strong>01. Regional modelling of present,<br />
past and future potential distribution of discontinuous permafrost<br />
based on a rock glacier inventory in the Bagnes-Heremence area<br />
(Western Swiss Alps). Norsk geog. Tidsskr. 55: 219-223.<br />
Sailer, R. and Kerschner, H. 1999. Equilibriurn-line altitudes and rock<br />
glaciers during the Younger Dryas cooling event, Ferwall group,<br />
western Tyrol, Austria. Annals of Glaciology 28: 141-145.<br />
Figure 2. Movement (direction and rate of displacement) of marked<br />
boulders (1-<strong>25</strong>) on the active rock glacier located in Val di Genova<br />
(A37 in the inventory). Sl=basis of topographic survey; S2=bench<br />
mark of topographic survey.<br />
The minimum altitude of activelinactive rock glaciers is<br />
utilized as an indicator of the present lower boundary of<br />
discontinuous permafrost (Barsch 1996), whereas the elevation<br />
of relict forms is used as an indicator of past permafrost<br />
conditions. In the study area the difference in altitude<br />
between the fronts of the activelinactive and relict<br />
forms indicates a increase in the lower boundary of discontinuous<br />
permafrost of 4<strong>20</strong> m. In other alpine areas, the<br />
relict rock glaciers have been ascribed to the Late Glacial<br />
(Sailer and Kerschner 1999, Frauenfelder et al. <strong>20</strong>01,<br />
Lambiel and Reynard <strong>20</strong>01). If the relict rock glaciers of<br />
the Adamello Presanella massif were tentatively correlated<br />
to Late Glacial by applying a mean temperature<br />
lapse rate of 0.65 "CIIOO m to the altitude variation of the<br />
lower limit of the discontinuous permafrost, a mean annual<br />
temperature rise of about 2.7 "C would be recorded.<br />
Among the activelinactive rock glaciers, about <strong>20</strong> are<br />
located in areas occupied by glaciers during the Little Ice<br />
Age (LIA) or are fed by glacial deposits from the same pe-<br />
146
8th International Conference<br />
Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Viable protozoa from permafrost sediments and buried soils<br />
A. V. Shatilovich, LAShmakova, S. V.Gubin and D.A. Gilichinsky<br />
Soil Cryology Laboratory, Institute of Physicochemical and Biological Problems in Soil Science, Russian Academy<br />
Pushchino, Moscow Region, Russia<br />
of Sciences,<br />
INTRODUCTION<br />
A significant number of viable ancient microorganisms is<br />
known to be present within permafrost. The age of the cells<br />
corresponds to the longevity of the permanently frozen state<br />
of the sediments, and the oldest date back to the Pliocene<br />
(Gilichinsky <strong>20</strong>02). Viable spores of the lower plants<br />
(mosses) and seeds of the higher plants were also found<br />
within Siberian and Canadian late Pleistocene permafrost; it<br />
was shown that they were still able to grow (Porsild et al.<br />
1967; Gilichinsky et al. <strong>20</strong>01; Gubin et al. <strong>20</strong>03). In order b<br />
extend our knowledge of biotic survival ongeologicaltime<br />
scales within permafrost, we aimed at the detection of viable<br />
protozoa. Protozoa represent a considerable part of soil biota<br />
and possess a wide spectrum of adaptation abilities.<br />
MATERIALS AND METHODS<br />
Figure 1. Geological section of icy-loess complex.<br />
The late Pleistocene icy-loess complex, Le., syngenetically<br />
frozen aleurites with the traces of soil (so-called cryopedolite), RESULTS AND DISCUSSION<br />
ice veins and buried soils on Kolyma lowland, has been<br />
studied. Cryopedolite contains up to 1% organic matter and Organic matter from burrows contain a large number of ele<br />
thin filamentous roots of grasses. Buried soils belong to His- ments from late-Pleistocene tundra-steppe ecosystems. This<br />
tosols and Gleysols. We also investigated samples from bur- great volume of paleoecological information survived due b<br />
rows of fossil gopher. The studied material consisted of mix- the frozen state of deposits from the time of their formation. Viture<br />
of aleurites with litter components: seeds, residues d able protozoa were found among other biological objects.<br />
leaves and grasses, wool of animals, etc. (Fig. I). Radiocar- Protists were isolated from all samples but one (collected<br />
bon age of burrows was determined as 28 to 32 thousand from the top layer of buried Histosol) and were identified as<br />
years (Gubin et al. <strong>20</strong>01).<br />
representatives of three large taxones: heterotrophic flagel-<br />
In order to find viable protozoa we used a method of soil lates, naked amoebae (Fig. 2), and small ciliates. Heterotroenrichmentcultivationonapoornutrientnon-selective<br />
me phic flagellates were dominant species of recovered commu-<br />
dium (sterile water and Lozina-Lozinskii medium). The cultivation<br />
was carried out during one month using two different<br />
temperatureAight conditions: 2OoC with daylight in one case<br />
and 5°C without light in the other. Protozoa were observed<br />
directly in the soil enrichment cultures and in suspension.<br />
Methods of light and fluorescent microscopy were used<br />
their morphological description and identification. images<br />
for<br />
d<br />
protists were obtained by means of a digital camera.<br />
nities. It was observed that a succession cycle (species<br />
composition change) of permafrost protests community takes<br />
approximately <strong>20</strong> days. A tendency appeared to exist for an<br />
increase in the number of viable protists and their species diversity<br />
in soil samples with high content of fossil plant . It<br />
was also observed that protozoa abundance and biodiversity<br />
was higher inside fossil burrows than in buried soils. This fad<br />
is likely to be due to initial relative abundance of protozoan<br />
fauna as well as to more favorable conditions of its cryopre-<br />
147
servation within the burrows. Following long-term survival in<br />
permafrost, the organisms become subjected to the stresses<br />
of thawing and exposure to oxygen as the samples are<br />
melted in the lab environment at relatively high temperatures.<br />
Becausethestrategiesandtechniquesforthe recovery d<br />
protozoa from frozen environments are just beginning to be<br />
developed, the standard methods for isolation of the protists in<br />
purecultureshavenotyetdeliveredpositiveresults. Most<br />
protozoa have not been cultured, and many of them died OT<br />
altered their activity.<br />
ACKNOWLEDGEMENTS<br />
This work was supported by the Russian Foundation of Basic<br />
<strong>Research</strong> (grants 01-05-65043 and 01-0449087).<br />
REFERENCES<br />
Gilichinsky, 0. Shatilovich A,, Vishnivetskaya T.., Spirina E., Faizutdinova<br />
R., Rivkina E., Gubin S., Erokhina L., Vorobyova E., Soina<br />
V. <strong>20</strong>01. How long the life might be preserved In (Giovannelli<br />
F., ed.), <strong>Proceedings</strong> of the International Workshop ((The<br />
Bridge between the Big Bang and Biology)), Consiglio Nazionale<br />
delle Ricerche of Italy: 362-369.<br />
Gilichinsky, D. <strong>20</strong>02. Permafrost as a microbial habitat I/ Encyclopedia<br />
of Environmental Microbiology, <strong>20</strong>02, 932-956.<br />
Gubin, S.V. et ai. <strong>20</strong>01. Composition of seeds from fossil gopher<br />
burrows in the ice-loess deposits of Zelyony mys asenvironmental<br />
indicator. Earth Cryosphere 5(2): 76-82 (in Russian).<br />
Gubin, S.V. et al. <strong>20</strong>03. The Possible Contribution of Late Pleistocene<br />
Biota to Biodiversity in Present Permafrost Zone /I Journal<br />
of General Biology <strong>20</strong>03, v.64 -1: 45-50 (in Russian).<br />
Porsild, A. et al. 1967. Grown seeds of Pleistocene age /I Science, v.<br />
158: 113-114.<br />
a.<br />
CONCLUSION<br />
For the first time, viable protozoa were discovered in frozen<br />
layers and their ability to survive within permafrost during<br />
dozens of thousands of years was shown. This suggests that<br />
in such a unique physicochemical environment as permafrost,<br />
these organisms keep unknown possibilities of physiological<br />
and biochemical adaptation incomparably longer than<br />
in any other known habitat. Their ability to survive on gee<br />
logical time scales within the broad limits of natural systems<br />
forces us to redefine the spatial and temporal limits of the terrestrial<br />
biosphere. On the basis of the obtained results, it is<br />
possible to propose the existence of viable protozoa not only<br />
in buried soils, but also in deposits rich in organic matter.<br />
Permafrost thawing due to natural or anthropogenic processes<br />
exposes ancient life to modem ecosystems and the ancient<br />
protozoa renew their physiological activity and participate in<br />
formation of modern biodiversity on the northern territories.<br />
148
Stress-strain conditions and the stability of arctic coastal slopes:<br />
assessments using seismic surveys<br />
A. G. Skvortsov and D.S. Drozdov<br />
Earth Cryosphere Institute, SB RAS, Russia<br />
Within the framework of ACD and INTAS-2332 projects,<br />
seismic research for the purpose of slope stability analysis<br />
was carried out at the Bolvanskii key-site in the mouth of<br />
Pechora river (Barents sea). A special seismic technique<br />
for investigations at various types of slopes, mainly at<br />
sliding slopes, was used. It is based on theoretical relationships<br />
and experimental correlation between seismic<br />
characteristics of the ground and stress-strain conditions<br />
of the soil. A very important feature of this technique is the<br />
possibility to monitor and predict preliminary changes of<br />
slope stability before discontinuity occurs. The term of forecast<br />
ranges from some months to several years. The<br />
uncertainty in spatial allocation of expected discontinuities<br />
does not exceed a few meters (Gorjainov et al. 1987,<br />
Skvortsov 1989).<br />
In previous years, the most adequate results were obtained<br />
for different sites in temperate climatic zones. The<br />
geological conditions of some examined sites can be considered<br />
to be a model of Arctic frozen slopes in summer<br />
time. The most interesting data was gained while investigating<br />
a coastal slope where limestone underlies a layer<br />
of clay. The processing of seismic data showed that the<br />
stable (for the date of research) part of the slope is going<br />
to be involved in the sliding process. Mainly the seismic<br />
data on the upper horizon of the clay layer were used to<br />
trace the position of the predicted crack (Alyoshin et al.<br />
<strong>20</strong>02). Within a few years, new landslides took place according<br />
to the forecast.<br />
The results of this research and numerous other experiments<br />
have shown that reliable information on the<br />
stress-strain conditions in slopes can be received on the<br />
basis of data on compressional wave velocity and its anisotropy<br />
in the uppermost part of a geological section (1 m<br />
thickness). From the methodical point of view this conclusion<br />
is extremely important for assessments of arctic<br />
coastal slope stability, since it shows that the stress-strain<br />
conditions of the coastal massif can be examined using<br />
the distribution of seismic characteristics in the active<br />
layer (seasonally-thawed layer).<br />
This conclusion was taken into account while carrying<br />
out in-situ field investigations at the northern slope of cape<br />
Bolvanskii where sea water undermines the third marine<br />
terrace (September, <strong>20</strong>02). Three main types of geological<br />
cross-sections are observed along the coast: 1) mainly<br />
peaty in top part; 2) mainly sandy; 3) mainly loamy. The<br />
loams contain many boulders. All soils are permanently<br />
frozen. Generally, they have relatively low ice-content (no<br />
more than 0.2). The increased ice-content (up to 0.3-0.4)<br />
is a characteristic of the upper 2-3 m of loamy crosssections.<br />
The peat bogs are the most ice-rich sites which<br />
include thick ice wedges (Malkova et al. <strong>20</strong>02).<br />
At the seismic key-site sand prevails in cross-section.<br />
The height of the coastal slope is about <strong>25</strong> m and the<br />
slope is rather abrupt (30"). However, any modern active<br />
processes, including landslides, crumbling and slumping,<br />
are not observed now while at nearby sites all these processes<br />
occur. Dense bushes, as seen on Figure 1, prove<br />
the modern stability of the investigated slope.<br />
The seismic operations were executed at four 50-60 m<br />
long profiles located perpendicularly to the coastal line.<br />
The profiles cross a sandy surface stretched along a ledge<br />
(bright grey band on Fig. I) and enter a peatbog (grey<br />
color). Information about the distribution of compressional<br />
and shear wave velocity, both in the active layer and in the<br />
top part of permanently frozen ground, was obtained. The<br />
preliminarily processing of these results showed obviously<br />
that the wave velocity greatly depends not only on the<br />
stress-strain distribution and further expected disturbances,<br />
but also on the variability of lithology, moisture,<br />
density, ice and organic content, etc.<br />
Other geophysical parameters depend less on these<br />
factors: the lateral anisotropy of wave velocity, Poisson's<br />
ratio and its anisotropy. To obtain them the wave velocity<br />
measurements were carried out in lateral rectangular directions<br />
(Skvortsov and Drozdov <strong>20</strong>02).<br />
The analysis of seismic characteristics of the seasonally<br />
thawed ground allows to define two weakened zones<br />
in the active layer. Most obvious, these zones are seen at<br />
the isoline maps of compressional wave velocity anisotropy<br />
and of Poisson's ratio anisotropy (Fig. 2).<br />
The first more sharply expressed weakened zone is<br />
situated in the north-western part of the site. It is subparallel<br />
to the ledge of terrace and is indicated as a zone<br />
with minimum values of wave velocity anisotropy and<br />
Poisson's ratio anisotropy.<br />
149
~igu~e 1. ~ ~ ~rQfiles ~ at the olvanskii s ~ ke~-sit~ i ~<br />
the ~ ~ ~ a~ i a n~ e ~ r~<br />
a o ~ ~ ~ ~ i n ~ .<br />
Skvortsov, A. G. et al.: Stress-strain conditions and the stability of ...<br />
and the ~ e~ult~ of<br />
An attempt has been undertaken to detect the presence<br />
of landscape attributes of this zone at the surface.<br />
For this purpose micro-landscape mapping was carried<br />
out during which an existing crack in the ground surface<br />
has been found <strong>20</strong> m to the west of the key-site in the first<br />
zone. That proves the first zone to be a weakened one,<br />
The first weakened zone indicates a line along which the<br />
separation of relatively small frozen blocks can move<br />
(“expected crack at Figures 1 and 2).<br />
The second weakened zone shows less contrast on<br />
isoline maps than the first zone. It is situated at <strong>20</strong>-40<br />
meters from the slope, crosses the first zone and ledge<br />
and probably is an attribute of larger frozen blocks which<br />
can be separated in a comparatively longer future. (“anisotropic<br />
ground strain” on Fig. I and 2).<br />
The submitted results illustrate a possibility of surface<br />
seismic techniques to investigate slope processes, to assess<br />
the stability of frozen ground and make spatio - tem-<br />
poral forecasts of coastal retreat at particular sites in the<br />
Arctic.<br />
ACKNOWLEDGEMENTS<br />
This project is supported by INTAS (grant 01-2332) and<br />
SB RAS grant.<br />
The authors express gratitude for cooperation and assistance<br />
with carrying out in-situ field investigations to:<br />
Norwegian Geotechnical Institute (manager of Oil Pollu-<br />
tion project 0. Nerland) and Circumpolar Active Layer<br />
Monitoring project (CALM, manager on European Russian<br />
sites G.V. Malkova).<br />
REFERENCES<br />
Alyoshin, AS., Drozdov, D.S., Dubovskoj, V.B. and Skvortsov, A.G.<br />
<strong>20</strong>02. New opportunities of geophysical monitoring of processes<br />
slope. Sergeev reading, issue 4. - Moscow; GEOS: 485-488 (in<br />
Russian).<br />
Gorjainov N.N., Bogolyubov, N.M., Varlamov, A.N., Matveev, V.S.,<br />
Nikitin, V.N. and Skvortsov, A.G. 1987. Studying of landslides by<br />
geophysical methods - Moscow; Nedra: 157 (in Russian).<br />
Malkova (Ananjeva), G.V., Drozdov, D.S., Kanevskiy, M.Z. and<br />
Korostelev, Yu.V. <strong>20</strong>02. The coastal researchs in the area of<br />
geocryological station “Cape Bolvanskiy ‘I, the estuary of Pechora<br />
river. 3-rd Workshop on Arctic Coastal Dynamics. - Oslo, IPA:<br />
27-28.<br />
Skvortsov, A.G. 1989. Monitoring of sliding slopes stability by use of<br />
seismic methods. Investig. of hydro-geol. and engineer.-geol. objects<br />
by geophysical and isotope methods. - Moscow: VSEG-<br />
INGEO: 3243 (in Russian).<br />
Skvortsov, A.G. and Drozdov, D.S. <strong>20</strong>02. The assessment of stressstrain<br />
conditions of coastal slope by use of seismoreconnaissance<br />
(seismic-service). 3-rd Workshop an Arctic Coastal Dynamics. -<br />
Oslo. IPA: 48.<br />
150
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Contraction crack dynamics in polygonal patterned ground and soil<br />
inflation in the Dry Valleys, Antarctica<br />
R.S. Sleften and B. Hallet<br />
Quaternary <strong>Research</strong> Center, University of Washington Seattle, WA, USA<br />
INTRODUCTION<br />
The Dry Valleys of Antarctica are believed to be part of an<br />
ancient landscape that has remained undisturbed for millions<br />
of years. Our recent work suggests, however, that<br />
the soils and land surfaces are relatively active due in part<br />
to seasonal opening and partial closure of contraction<br />
cracks that delineate the pervasive polygonal patterned<br />
ground in this region (PBwB 1959). Unlike the relatively<br />
moist Arctic where contraction cracks tend to fill with surface<br />
meltwater and form ice wedges, contraction cracks in<br />
the Dry Valleys tend to fill with sand and form sand<br />
wedges. The growth of sand wedges disturbs the ground<br />
surface and the development of soil profile. We examined<br />
the rates of sand wedge growth by continuing a four decade-long<br />
survey of displacements across contraction<br />
cracks, and by installing modern instrumentation.<br />
RESULTS AND DISCUSSION<br />
To measure the growth of sand wedges, Robert F. Black<br />
and Thomas Berg (Berg and Black 1966; Black 1973) installed<br />
over 500 pairs of 60cm vertical rods that span contraction<br />
cracks in the Dry Valleys during the early to mid<br />
1960s. They measured crack divergence manually using<br />
precision rulers. We revisited and measured Black’s 13<br />
sites on three occasions since 1995; we found the rods to<br />
be generally in good condition and that crack divergence<br />
continued at rates (0.4-2 mm a”) similar to those measured<br />
by Black (Figure 1).<br />
To provide a better understanding and a basis for<br />
modeling the annual dynamics of contraction cracks, we<br />
installed linear transducers connected to data loggers at 3<br />
sites in Beacon Valley and Victoria Valley in 1998, and<br />
have continuously monitored displacements across cracks<br />
since that time. In addition, we monitor the micrometerology,<br />
soil temperature, soil thermal properties, and other<br />
physical parameters.<br />
The total annual crack opening due to thermal contraction<br />
of polygons ranges from 0.5 cm to nearly 2 cm at<br />
other sites, and lags air temperature (figure 2). The con-<br />
30 I<br />
<strong>25</strong><br />
<strong>20</strong><br />
15<br />
10<br />
5<br />
0<br />
n<br />
0 IO <strong>20</strong> 30 40<br />
-3-<br />
3 Time since (years) installation<br />
v<br />
Figure 1, Average contraction crack growth as determined by rods installed<br />
at 3 Black’s sites and measured manually. Initial and 7-year<br />
measurements were made by Black. Our measurements were made<br />
35 and 39 years after the rods were pounded into the permafrost.<br />
traction cracks open during cooling as the ice-rich ground<br />
contracts and expands as the ground warms. The net annual<br />
crack divergence reflects the amount of fines that fall<br />
into the cracks thereby preventing full closure. The 40-<br />
year average rate of divergence across the cracks based<br />
on Black‘s rods is 0.6 mm a-1 for the lower Beacon Valley<br />
site, whereas our continuous measurements for the past 4<br />
years show a net crack divergence rate of 1.4 mm a”. If<br />
the more rapid recent growth is significant, it presumably<br />
reflects an increase in either the amplitude of the thermal<br />
forcing or the sediment input into cracks.<br />
The long-term growth of sand wedges has caused an<br />
effective inflation of the soils and, in lower Beacon Valley,<br />
raised the ground surface by at least 8 m, the depth of our<br />
deepest borehole there. Preferential injection of finegrained<br />
material to depth, as the larger rocks do not fit into<br />
the open contraction cracks, ultimately results in the accumulation<br />
of rocks and boulders on the surface. The<br />
sand wedges grow to encompass the entire surface and<br />
appear to repeatedly sweep across the surface, thereby<br />
building up a complex of sand wedges overlain by rocks.
Sletten, R.S. et al.: Contraction crack dynamics in polygonal patterned ground<br />
REFERENCES<br />
<strong>20</strong><br />
0<br />
-<strong>20</strong><br />
-40<br />
<br />
3<br />
8 -60<br />
6 16<br />
3<br />
h<br />
14<br />
".<br />
Y<br />
12<br />
10<br />
8<br />
6<br />
4<br />
0<br />
$ 2<br />
d<br />
% - O<br />
8 1199 7199 1/00 7100 1/01 7101 1102 7102 1103<br />
"-.<br />
3 Date<br />
Y<br />
Figure 2. Continuous record of surface temperature and divergence<br />
across contraction cracks at the periphery of sand-wedge polygons<br />
using linear transducers mounted on rods spanning the cracks. The<br />
dashed line indicates a net annual divergence rate of 1.4 mm a.1.<br />
Berg, T. E. and Black, R. F. 1966. Preliminary measurements of<br />
growth of nonsorted polygons, Victoria Land, Antarctica. Antarctic<br />
Soils and Soil Forming Processes, Antarctic <strong>Research</strong> Series, Vol.<br />
8. J. C. F. Tedrow. American Geophysical Union, Washington,<br />
DC: 61-108.<br />
Black, R. F. 1973. Growth of patterned ground in Victoria Land, Antarctica.<br />
Permafrost: The North American Contribution to the 2nd<br />
International Conference, Yakutsk, Siberia, National Academy of<br />
Sciences: 193-<strong>20</strong>3.<br />
Schirrmeister, L., Sieged, C. Kuznetsova, T. Kuzmina, S. Andreev, A.<br />
Kienast, F. Meyer, H. and Bobrov, A. <strong>20</strong>02. Paleoenvironmental<br />
and paleoclimatic records from permafrost deposits in the Arctic<br />
region of Northern Siberia. Quaternary International 89(1): 97-118.<br />
Pewe, T. L. 1959. Sand-wedge polygons (Tesselations) in the<br />
McMurdo Sound region, Antarctica-a progress report. American<br />
Journal of Science <strong>25</strong>7: 545-552.<br />
This injection of fine-grained sediment into contraction<br />
cracks and the resulting inflation of the surface in the Dry<br />
Valleys appear analogous to the formation of the ice complexes<br />
in Siberia due to the recurrent build-up of ice in<br />
contraction cracks (Schirrmeister et al. <strong>20</strong>02). In both<br />
cases, soil inflation has been occurring for long periods.<br />
Our preliminary cosmogenic isotope data indicate that<br />
sand from sand wedges at 5 m depth has been buried for<br />
several million years, suggesting long-term burial and inflation<br />
and, possibly, shielding by the Taylor Glacier during<br />
a former excursion into Beacon Valley.<br />
CONCLUSIONS<br />
Current rates of sand-wedge growth (-1 mm a") and<br />
polygon sizes (-10 m) suggest that the near-surface layer<br />
of polygons would be completely resurfaced, in a period of<br />
tens of thousands of years, while rocks remain at the surface<br />
indefinitely as sand-wedges grow beneath them. The<br />
continuous injection of fines into contraction cracks has<br />
created a sand wedge complex that has inflated the surface<br />
over 8m. Our findings suggest that landscape surfaces<br />
with well-developed polygonal patterned ground in<br />
the Dry Valleys are relatively active, which contrasts with<br />
suggestions that these surfaces have been undisturbed<br />
for millions of years.<br />
152
The Global Terrestrial Network for Permafrost (GTN-P) - status and<br />
preliminary results of the thermal monitoring component<br />
S.L. Smith, M. M. Burgess, *V. Romanovsky and **J. Brown<br />
Geological Survey of Canada, Natural Resources Canada<br />
*University of Alaska, Fairbanks, USA<br />
**International Permafrost Association, Woods Hole, Mass., USA<br />
INTRODUCTION<br />
The Global Terrestrial Network for Permafrost (GTN-P) was<br />
established in 1999 to provide long-term field observations of<br />
active layer and permafrost thermal state that are required b<br />
determine the present permafrost conditions and to detect<br />
changes in permafrost stability. The data supplied by this<br />
network will enhance our ability to predict the consequences<br />
of permafrost degradation and to develop adaptation measures<br />
to respond to these changes. The GTN-P contributes to the<br />
World Meteorological Organization's Global Climate Observing<br />
System and Global Terrestrial Observing System.<br />
An overview of the GTN-P is given by Burgess et al.<br />
(<strong>20</strong>00).<br />
The active layer component of the GTN-PI the Circumpolar<br />
Active Layer Monitoring (CALM) program, has been in<br />
operation over the last decade. Organizational efforts of the<br />
GTN-P are therefore focused on the recently established<br />
borehole thermal monitoring component. This paper describes<br />
the present status of the thermal monitoring component and<br />
presents preliminary results from North America.<br />
STATUS OF THE THERMAL MONITORING<br />
PROGRAM<br />
The permafrost thermal monitoring program consists of a<br />
globally comprehensive network of boreholes for ground temperature<br />
measurements. Many of these boreholes were<br />
drilled for research, geotechnical, or resource exploration purposes<br />
in the last two decades and have been maintained as<br />
thermal monitoring sites. About 370 boreholes from 16 countries<br />
have been identified as candidate sites for inclusion in<br />
the GTN-P thermal monitoring system (Fig. I). The majority<br />
of these boreholes are between 10 and 1<strong>25</strong> meters deep and<br />
areinthenorthemhemisphere.A few sitesarelocatedin<br />
Antarctica and Argentina. Regional networks include those of<br />
the Geological Survey of Canada (GSC) in the Mackenzie<br />
region, the University of Alaska's Alaskan transect, the<br />
United States Geological Survey's deep boreholes in Northem<br />
Alaska and the European Community's Permafrost and<br />
Climate in Europe (PACE) program of boreholes largely in<br />
alpine permafrost. A web site (http://sts.qsc,nrcan.qc.ca/atnp)<br />
developed by the GSC, provides an inventory of candidate<br />
boreholes, and background material on the GTN-P.<br />
Metadata compilation for the candidate boreholes has<br />
been conducted over the last two years. Metadata from about<br />
60% of the boreholes has been compiled and posted on the<br />
ETN-P website. Over the next year, metadatacompilation<br />
will be completed and site selection will take place. Compilation<br />
of summary historical data for network sites will also occur<br />
over the next year.<br />
PRELIMINARY RESULTS<br />
A general discussion of preliminary results from GTN-P sites<br />
canbefound in Romanovsky et al. (<strong>20</strong>02). Thispaper b<br />
cuses on results from across the North American Arctic.<br />
Analysis of permafrost temperatures, taken in theupper<br />
<strong>20</strong> m, from a network of boreholes in Alaska maintained by<br />
the University of Alaska Fairbanks indicates that permafrost<br />
has warmed over the last <strong>20</strong> to 50 years. At the recently reactivated<br />
monitoring site at Barrow, temperatures at a depth d<br />
15 m increased 1°C between 1950 and <strong>20</strong>01. Increases of<br />
permafrost temperature ranging from 0.6 to 1.5"C between<br />
1983 and <strong>20</strong>00 at a depth of <strong>20</strong> m have been observed at<br />
sites along the Trans-Alaska pipeline route.<br />
The GSC has operated a network of thermalmonitoring<br />
sites in the Mackenzie region, Northwest Territories since<br />
1985. At 15 m depth at a site in the central Mackenzie valley<br />
an increase of about 0.03"C per year was observed be<br />
tween 1986 and <strong>20</strong>01 in permafrost with temperatures of ap<br />
proximately -1°C. In the northem Mackenzie region, in<br />
colder permafrost (-7"C), temperatures at a depth of 28 m<br />
show an increase of about 0.1 "C per year in the 1990s.<br />
At the GSC's high Arctic observatory at Alert, a warming<br />
of the permafrost occurred in the late 1990s of 0.15°C per<br />
year at a depth of 15 m. Some cooling of permafrost was<br />
observed from the late 1980s to the early 1990s in the eastern<br />
Canadian Arctic at a depth of 5 m at Environment Canada's<br />
borehole at Iqaluit. This cooling however, was following by<br />
warming in the late 1990s. This trend is similar to that<br />
observed in Northern Quebec, where cooling of about 0.1 "C<br />
153
Smith, S.L. et al.: The Global Terrestrial Network for Permafrost (GTN-P) ...<br />
Figure 1. Permafrost distribution in the Northern Hemisphere and location of candidate boreholes for permafrost thermal monitor-<br />
per year was observed from the mid 1980s to the mid 1990s<br />
at a depth of IOm in boreholes maintained by Universite Laval<br />
(Allard et al. 1995). This was followed by a warming<br />
trend starting in 1996 (Brown et <strong>20</strong>00). al.<br />
SUMMARY<br />
The GTN-P has been established to provide long-term ob<br />
servations of permafrost conditions that will enable us to address<br />
a number of climate change issues in the permafrost<br />
regions. The data collected by this network will provide information<br />
to scientists studying the cryosphere, and also to<br />
stakeholders, politicians and other decision makers. This information<br />
is critical in order to evaluate the impacts of climate<br />
change and to develop adaptation measures to reduce these<br />
impacts.<br />
Results from North America indicate that warming of permafrost<br />
occurred in the latter half of <strong>20</strong>th the century in Alaska<br />
and the western Canadian Arctic. Warming of permafrost is<br />
also observed in the Canadian eastem and high Arctic and<br />
mainly occurred in the late 1990s. These trends in permafrost<br />
temperature are consistent with the increases in air temperature<br />
observed since the 1970s in the North American Arctic.<br />
Brown, J., Hinkel, K.M. and Nelson, F.E. <strong>20</strong>00. The Circumpolar Active<br />
Layer (CALM) program: research designs and initial results.<br />
Polar Geography 24: 163-<strong>25</strong>8.<br />
Burgess, M.M., Smith, S.L., Brown, J., Romanovsky, V. and Hinkel, K<br />
<strong>20</strong>00. Global Terrestrial Network for Permafrost (GTNet-P):<br />
permafrost monitoring contributing to global climate observations.<br />
Geological Survey of Canada, Current <strong>Research</strong> <strong>20</strong>00-<br />
E14.<br />
Romanovsky, V. Burgess, M., Smith, S., Yoshikawa, K., and Brown, J.<br />
<strong>20</strong>02. Permafrost temperature records: indicators of climate<br />
change. EOS, Transactions of the American Geophysical Union<br />
83 (50): 589.<br />
REFERENCES<br />
Allard, M., Wang, B. and Pilon, J.A. 1995. Recent cooling along the<br />
southern shore of Hudson Strait Quebec, Canada, documented<br />
from permafrost temperature measurements, Arctic and Alpine<br />
<strong>Research</strong> 27: 157-1 66.<br />
154
8th international Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Satellite radar interferometry for detecting and quantifying mountain<br />
permafrost creep<br />
T. Strozzi, U. Wegmiiller, *A. Kaab and R. Frauenfelder<br />
Gamma Remote Sensing, Muri BE, <strong>Switzerland</strong><br />
"Glaciology and Geomorphodynamics Group, Department of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
Remote-sensing techniques represent suitable tools for<br />
integral hazard mapping and monitoring in high mountains,<br />
regions that are typically difficult to access. Spaceborne<br />
Synthetic Aperture Radar (SAR) systems offer the<br />
possibility, through differential SAR interferometry (Dln-<br />
SAR), to map surface displacements at mm to cm resolution<br />
in the line-of-sight direction (Bamler and Hart1 1998).<br />
Recent studies on rock glacier deformation suggest that<br />
the technique is promising also for relatively small objects<br />
in mountainous terrain (Rignot et al. <strong>20</strong>02; Kenyi and<br />
Kaufmann, <strong>20</strong>03).<br />
We analyzed a series of SAR images between 1993<br />
and 1999 from the European Remote Sensing Satellites<br />
ERS-I and ERS-2 (C-Band, 5.3 GHz, 5.6 cm wavelength)<br />
and the Japanese Earth Resources Satellite JERS-1 (L-<br />
Band, 1.3 GHz, 23.5 cm wavelength). From the large<br />
number of the possible interferometric pairs we considered<br />
interferograms with short baselines acquired during<br />
the snow-free period between early summer and mid-fall.<br />
The topographic reference was determined from two winter<br />
ERS-112 Tandem pairs and subtracted from the interferograms.<br />
The differential interferograms were orthorectified<br />
to the Swiss cartographic system with a pixel spacing<br />
of 10 m by use of a Digital Elevation Model (DEM) derived<br />
from aerial photogrammetry (Kaab <strong>20</strong>02). For the most<br />
convincing interferograms the analysis continued with<br />
phase unwrapping permitting the quantitative measurement<br />
of the displacement field of identified rock glaciers.<br />
Phase unwrapping of interferograms in rugged terrain and<br />
for complicated displacement fields is a critical task that<br />
was successful only for small areas. The line-of-sight displacement<br />
was finally transformed in displacement along<br />
the slope direction using the DEM.<br />
Results for the Fletschhorn region in the SimplonlSaas<br />
valley region, Swiss Alps, were analyzed together with the<br />
survey of rock glaciers (Frauenfelder et al. 1998) and displacement<br />
maps from digital photogrammetry of repeated<br />
airborne imagery (Kaab <strong>20</strong>02). Two examples are shown<br />
in Figures 1 and 2. For large active rock glaciers such as<br />
the Rothorn (Figure 1) JERS DlnSAR is able to display the<br />
entire flow field with the typical velocity distribution of<br />
highest speeds in the rock glacier center. Below the Jegi<br />
rock glacier (Figure 2) the advantage of DlnSAR for detecting<br />
small movements becomes apparent. Field classification<br />
gave a transition from active over inactive to relict<br />
parts. In the ERS differential interferogram surface deformation<br />
can clearly be seen. It is however an open issue<br />
whether deformations observed from DlnSAR for the inactive<br />
and relict parts of the rock glaciers are due to 'active'<br />
ice deformation or due to 'passive' pushing by the upper<br />
active rock glaciers. Also for the debris slope to the south<br />
of the Jegihorn, which was not classified as rock glacier, it<br />
is not clear if the small movements are due to deformation<br />
of frozen debris.<br />
For rock glacier research and assessment of related<br />
slope instability hazards, our study suggests the following<br />
prior fields of application for DlnSAR: (i) DlnSAR is especially<br />
useful for area-wide detection of rock glacier activity<br />
and order of surface velocity; (ii) the detection of very<br />
small movements by DlnSAR is feasible and will provide<br />
new insights into the behavior of inactive or even relict<br />
rock glaciers; (iii) such results reveal valuable parameters<br />
for further detailed investigations at selected sites by<br />
photogrammetric or field-based methods; (iv) DlnSAR is<br />
able to provide rock glacier speed for a large number of<br />
such landforms and might promote the understanding of<br />
the coupling between landscape, climate and rock glacier<br />
activity.<br />
The major limiting factors of DlnSAR in mountainous<br />
terrain arise from temporal decorrelation and the SAR image<br />
geometry, both leading to incomplete spatial coverage.<br />
Over rock glaciers, where dense vegetation is no<br />
longer present, high coherence is regularly observed during<br />
the snow-free period. Also, large and incoherent displacements<br />
of adjacent scatters may cause decorrelation.<br />
Coherence maps may therefore support the detection of<br />
displacements. The very rugged topography of alpine regions<br />
causes incomplete coverage due to layover and<br />
shadowing. Furthermore, there is a privileged slope direction,<br />
namely that facing away from the SAR look vector,<br />
where the technique is better suited for the monitoring of<br />
displacements.<br />
155
tory afound The Inner ~ o~ho~n from in-situ<br />
nt v ~ l ~ ~ c ~ i ~ from ~ i vsa~~i!ite<br />
e ~<br />
d ~ ma~e is an a ~ r i ~ l i ~ ~ ~ e r y ~<br />
indi~ato~s, such as ~eomorphoio~i~al ~nter~ret~tion~ ~ ~ wa- ~<br />
on of vege~~tion cover^ and ioeation~ of burrows of ~ ib~rna~<br />
, in~€~iv~<br />
dark gr~y), and relict ( l i gray^. ~ ~ Image ~ wid^^ is 7.5<br />
€ui~ition time in~e~al and -65 m perp~n~i~ular ba~eli (c)<br />
ming creep in the direet~on of maximum slope (max~mum 60<br />
the ~e~ihorn from in-situ na~u~al indi€~tors~ ~egend as in Fi~ur~<br />
I. Image width is 1.5 km. (b) ERS<br />
'10.2~ with 105 days a€ui~ition time inte~al and 5 m perpen~i~ular bas~lin~. (c) ~u~~<br />
ffom sate!ii~e radar in~e~e~~metry by assuming creep in the di~~~tion of maximum lo^^ (maxi~um 10 ~ m/y~~r).<br />
Rignot, E., Hallet, B., and Fountain, A. <strong>20</strong>02. Rock glacier surface<br />
ACKNOWLEDGEMENTS<br />
motion in Beacon Valley, Antarctica, from synthetic-aperture radar<br />
ERS SAR data courtesy A03-178,O ESA, processing GAMMA. JERS<br />
interferometry. Geophysical <strong>Research</strong> Letters,<br />
GL013494,29 June <strong>20</strong>02.<br />
10.10291<strong>20</strong>01<br />
SAR data courtesy J-2RI-001, 0 NASDA, processing GAMMA. Aerial<br />
imagery taken by Swisstopo and the Swiss Federal Office of Cadastral<br />
Surveys.<br />
REFERENCES<br />
Bamler, R., and Hartl, P. 1998. Synthetic aperture radar interferometry.<br />
Inverse Problems, 14, RI-R54.<br />
Frauenfelder, R., Allgower, B., Haeberli, W., and Hoelzle, M. 1998.<br />
Permafrost investigations with GIS - a case study in the<br />
Fletschhorn area, Wallis, Swiss Alps. Proc. 7th Intern. Conf. Permafrost,<br />
Yellowknife, Canada, Collection Nordicana. 57: 291-295.<br />
Kaab <strong>20</strong>02. Monitoring high-mountain terrain deformation from digital<br />
aerial imagery and ASTER data. ISPRS Journal of Photogrammetry<br />
and remote sensing. 57(1-2): 39-52.<br />
Kenyi, L.W. and Kaufmann, V. <strong>20</strong>03. Measuring rock glacier surface<br />
deformation using SAR interferometry. 8th Int. Conference on<br />
Permafrost. <strong>Zurich</strong>.<br />
156
8th International Conference an Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
An attempt to obtain paleoclimatic information from the present permafrost<br />
distribution in Siberia<br />
T. Sueyoshi and *Y. Hamano<br />
Laboratory of Hydraulics, Hydrology and Glaciology, ETH <strong>Zurich</strong><br />
*Department of Earth and Planetary Sciences, Univ. of Tokyo, Japan<br />
INTRODUCTION<br />
The thickness of permafrost changes in response to climate<br />
change. Changes in surface temperature are transmitted<br />
through thermal conduction, thus, the response time becomes<br />
long for thick permafrost.<br />
Siberia is a region in which some of the thickest permafrost<br />
is found, sometimes more than several hundreds meters<br />
deep. Such thicknesses of permafrost should reflect the surface<br />
temperature history of the past. In spite of its importance<br />
for the understanding of long-term climatic variability, the terrestrial<br />
paleotemperature history of the continental inland is still<br />
open to argument because most "high-resolution" paleoclimatic<br />
studies are based on marine sediments or ice cores. In<br />
this study, we attempt to obtain paleoclimatic information from<br />
the present permafrost distribution in Siberia by means d<br />
numerical methods.<br />
METHOD<br />
We used a one-dimensional numerical model of permafrost in<br />
which the ground temperature profile is calculated in response<br />
to the variations of surface temperature change from LGM<br />
(Last Glacial Maximum, 21 ka) to the present.<br />
We worked with the one-dimensional thermal conduction<br />
problem, we took latent heat production into account but ne<br />
glected soil water diffusion. Latent heat was considered by<br />
using the enthalpy method, which generally gives good results<br />
when calculating frozenlunfrozen boundary positions.<br />
The thickness of the permafrost is obtained by tracking the<br />
position of this frozen front.<br />
Three study sites for numerical experiments, Tiksi (71"N),<br />
Yakutsk (62'N) and lrkutsk (52"N) were selected in this<br />
study to investigate the paleoclimatic variability in east Siberia.<br />
These sites are located along the Lena River and have<br />
similar geological conditions. The locations, permafrost thickness<br />
and the mean annual air temperature (MAAT) of the<br />
sites are shown in table 1.<br />
Table 1.<br />
Site name Lat. Log thickness MAAT<br />
Tiksi 71.35'N 128.55'E 650m -14'C<br />
Yakutsk 62.05'N 129.45'E 500m- -8KC<br />
lrkutsk 52.16'N 104.21"E <strong>25</strong>m- -0.OT<br />
These permafrost thicknesses are based on<br />
and Devyatkin (1974) and Harada (<strong>20</strong>01).<br />
Nekrasov<br />
Boundary Conditions:<br />
Conditions considered in this simulation are summarized as<br />
follows:<br />
I) Geothermal heafflow value (30 - 60 mW/m2)<br />
2) Local climate condition:<br />
annual mean temperature, seasonal variability<br />
3) Temperature drop during LGM (DT)<br />
a) "normal" paleocliamte scenario:<br />
same as the ice-core reconstruction (DT-10'C)<br />
b) "extreme" paleocliamte scenario:<br />
colder climate (DT -15'C, <strong>20</strong>"C, <strong>25</strong>°C)<br />
4) Effect of ice sheet existence (only in the Tiksi case)<br />
Top and bottom boundary conditions are necessary to tun<br />
the model. Geothermal heatflow was given for the bottom<br />
boundary conditions and surface temperature was used for<br />
the top.<br />
We assumed the value of the heatflow to be constant for<br />
the calculated period since the study area is on a stable<br />
shield continent. We found this assumption to be reasonable.<br />
Surface ground temperature is substituted by air temperature.<br />
Present climatic data was used when considering the<br />
climatic conditions of the study sites. Paleoclimate history<br />
was obtained from reconstructed temperaturedeviation data<br />
from Greenland ice cores (GRIP, 1993) in order to represent<br />
the outline of climate change. The general pattern of climate<br />
change was assumed to be basically the same for all sites,<br />
while the temperature drop between Holocene and LGM was<br />
treated as a parameter. The blanketing effect of the ice sheet<br />
during the glacial period was also considered.<br />
The effect of the ice sheet on ground temperature was als<br />
taken into account in the model. The ground temperature un-<br />
157
der the ice sheet was calculated from the air temperature d<br />
the surface of ice sheet and assuming a temperature gradient<br />
through the ice sheet. The air temperature on the of top the ice<br />
sheet was calculated using the adiabatic gradient of the atmosphere<br />
(0.6"C/IOOm) and assuming the thermal conductivity<br />
of ice to be 2.2 [Wlklm]. The thickness of the ice sheet<br />
was also treated as a parameter but this condition was considered<br />
only for the northSiberian case. The reconstruction d<br />
ice sheet distribution was based on CLIMAP (1976) and<br />
Pertier (1994).<br />
Sueyoshi, T. et aL: An attempt to obtain paleoclimatic information ...<br />
ness around Tiksi can be reproduced by the "normal" palm<br />
temperature scenario, thereby this "extreme" cooling is<br />
thought to be characteristic for inland Siberia.<br />
1<strong>20</strong>0 , , I I I I I I I ( I . I .<br />
I<br />
I . . . . , . , * , ....<br />
Yakutsk (62'N, 1 JO'E)<br />
RESULTS AND DISCUSSIONS<br />
the<br />
A series of numerical experiments was performed under<br />
various boundary conditions mentioned in the previous section.<br />
We obtained the temporal variations of permafrost thickness<br />
for the last 30k years (from LGM to the present).<br />
0 5000 10000 15000 <strong>20</strong>000 <strong>25</strong>000 30000<br />
In this paper, we show the results from the parameter<br />
yearsBP<br />
study of the temperature drop during LGM. The target of these Figure 2. Temporal change of permafrost thickness at Yalutsk.<br />
experiments was to examine the sensitivity of permafrost<br />
thickness to the climate during LGM and to argue to what On the other hand, the distribution of permafrost in the<br />
extent the present thickness of the permafrost has been af- south-marginal area could not be reproduced, even under<br />
fected by past climate history.<br />
"extreme" conditions. Since such thin permafrost responds<br />
Figure1 shows the temporal variation of the permafrost quickly to surface climate change it should not be regarded as<br />
thickness around Tiksi, from 30ka to the present, using sev- a relic of a past cold climate.<br />
eral different values of temperature drop during LGM (DT). In<br />
this plot the X-axis represent the years before present while<br />
the Y-axis represents the thickness of the permafrost. The ar- CONCLUSIONS<br />
row at the Y-axis shows the present thickness of permafrost<br />
at that location. From the figure we can tell that the present The permafrost thickness in the central (Yakutsk) and northem<br />
permafrost thickness in this region roughly agrees with the part (Tiksi) of Siberia cannot be explained consistently by the<br />
"normal" paleoclimate scenario.<br />
same paleoclimatic boundary conditions. The present pemafrost<br />
thickness at Tiksi generally agrees with the value of the<br />
Tiksl(7laN,1 29'E)<br />
1<strong>20</strong>0,....,.., . . . . . . . . . . . . . . . . . . . . . .<br />
calculated thickness, while that at Yakustk requires colder<br />
- 1000<br />
E<br />
Y<br />
U<br />
IC<br />
E<br />
800<br />
0 600<br />
n<br />
c<br />
400<br />
i<br />
............ J. ...... 1 ....... !..".,_"<br />
..I<br />
--.-"'1'"' ! : AT 5 S'C<br />
............................ :. ............!... ... ................................. ......<br />
I<br />
o ~ " " ' " ' ~ ' " " ' " ' " ' ' " ' ' ~<br />
0 5000 10000 15000 <strong>20</strong>000 <strong>25</strong>000 30000<br />
yearsBP<br />
Figure 1. Temporal change of permafrost thickness at Tiksi.<br />
Figure 2 shows the case for Yakutsk in the central part of<br />
Siberia.<br />
Comparison with the present thickness suggests that this<br />
area could have experienced enhanced cooling during LGM<br />
which is different to that obtained from the reconstructed ice<br />
core data. If we do not consider this effect the conditions for<br />
this region are too mild, the thickness of permafrost should be<br />
thinner than it is at present. However, the permafrost thick-<br />
conditions. This suggests that the temperature drop during<br />
LGM in the central part of Siberia was larger in the north.<br />
REFERENCES<br />
CLIMAP Project Members. 1976. The Surface of the Ice-Age Earth;<br />
Science 191: 1131-1137.<br />
GRIP Members. 1993. Climate instability during the last interglacial<br />
period recorded in the GRIP ice core; Nature 364: <strong>20</strong>3-<strong>20</strong>7.<br />
Harada, K. <strong>20</strong>01, Study on detection of permafrost structure., DOCtoral<br />
Thesis in Hokkaido Univ.<br />
Nekrasov, LA. and Devyatkin, V.N. 1974. Morphology of permafrost -<br />
underlain areas of the Yana River basin and adjacent regions,<br />
Nauka Press, Novosibirsk (in Russian).<br />
Pertier, W.R. 1994. Ice Age Paleotopography, Science 265: 195-<br />
<strong>20</strong>1.<br />
158
8th International Conference on Permafrost Extended Abstracts on ~~~~~~~<br />
<strong>Research</strong> and Newly Available lnforrnatlan<br />
Isotopic composition of the massive ground iceof the Bovanenkovo<br />
geotectonic structure (Yamal Peninsula, Russia)<br />
A. M. Tarasov, 1. D. Streletskaya* and N. G. Ukraintseva*<br />
“GROUND” CO. LTD, Moscow, Russia<br />
*Moscow State University, Moscow, Russia<br />
INTRODUCTION<br />
A study of massive ice deposits was carried out within the<br />
framework of an engineering geo-cryological survey in the<br />
area of the Bovanenkovo gas-condensate field. Ten large<br />
ice beds, situated at a distance of 5 to 60 km from each<br />
other, were drilled from top to bottom (Kurfurst et al. 1993,<br />
Tarasov 1990, 1993). The ice and host deposits were<br />
sampled, with an increment of 10-15 cm, to determine<br />
their chemical and isotopic composition.<br />
RESULTS AND DISCUSSION<br />
Morphological features of the massive ground ice<br />
It has been established that the large ice deposits stretch<br />
for <strong>20</strong>0-800 m and that their thickness varies from 5 to<br />
more than 30 m. They form beds with a horizontal area<br />
that can exceed the vertical by up to a dozen times. Deposits<br />
of this type are widely spread in sections of the Upper<br />
Pleistocene marine plains and lie mainly in the contact<br />
zone between the saline marine clays (Ds=0.3-15%), and<br />
the slightly salinized (Ds=0.02-0.1%) marine-lacustrine<br />
sandy-silt sediments that lie at depths from 1-2 m to more<br />
than <strong>20</strong>-30 m (Dubikov <strong>20</strong>02). Massive ground ice bodies<br />
are more than 15 m in thickness. In all studied deposits<br />
the bottom of the ice beds is flat, sub-horizontal and lies<br />
beneath the modern sea level (Fig. 1). The contact is consistent<br />
with substrate strike. The ice top is rolling due to<br />
diagenetic processes (melting, thermodenudation and<br />
thermoerosion). Locally, the bedding is concordant with<br />
the covering clay sediments (coarse ice bars). However in<br />
some cases, when the ice is covered by solifluctionlandslide<br />
deposits with cryogenic structure, the contact is<br />
sharply discordant, gently sloping and the ice top is<br />
melted. Ice has been noted with local inclusion of salty<br />
water, cryopegs, voids, cracks and under-ice gases<br />
(Streletskaya and Leibman <strong>20</strong>02).<br />
Chemical and isotopic composition of the massive ground<br />
ice<br />
The chemical composition of massive ground ice was<br />
analyzed at study sites 1-88 (central part of the gas field<br />
area). An exposure of massive ground ice was found in a<br />
west slope relic hill in 1987. The bottom of the massive<br />
ground ice body is horizontal and situated at 1-3 m above<br />
sea level. The massive ground ice body is flat-layered, ultra<br />
fresh, and abundant in mineral inclusions. The salt<br />
content in pure ice (melted) is very low and varies from 10<br />
to 80 mgldm3. The salt content in massive ground ice with<br />
mineral inclusions increases up to 160-180 mgldm3.<br />
Water-soluble HC03- is the dominant anion and Na+, Catt<br />
dominated among cations.<br />
The oxygene-18 isotope content (6180) in massive ice<br />
melts is shown in table 1. The background measurements<br />
of 6180 are the same in all ice bodies even though they<br />
can be 30-50 km removed from each other. This proves<br />
that the all massive ground ice in this region has a common<br />
origin. Meteoric water was the source for the formation<br />
of this massive ground ice, marine water did not take<br />
part in this process.<br />
Table 1. Oxygene-18 isotope (8180) in massive ground ice.<br />
Study H a.s.l., Depth of 6180 6180 6180 n*<br />
Site1 m the rn ice, Mean Min Max<br />
Pnint<br />
1-88/2 6.8 1.2-8.7 -18.4 -17.7 -19.7 13<br />
1-8813 10.3 1.8-11.0 -18.0 -17,l -19.0 16<br />
1-8814 12.1 3.7-14.6 -18.4 -18.0 -19.0 <strong>20</strong><br />
1-8817 14.4 6.4-15.4 -18.5 -18.1 -19.0 7<br />
2-8811 21.9 6.1-12.0 -18.8 -18.0 -19.7 IO<br />
2-88162 6.4 1.7-7.5 -18.9 -18.6 -19.3 8<br />
2-89122 15.6 3.3-12.8 -18.9 -17.3-<strong>20</strong>.2 18<br />
3-8911 11.4 3.6-5.3 -19.7 -19.0 -<strong>20</strong>.6 4<br />
1-9013 7.0 1.0-15.3 -19.0 -17.7 -22.5 16<br />
2-90132 16.4 1.8-10.6 -18.8 -17.6 -<strong>20</strong>.6 11<br />
2-9011 17.0 1.2-13.2 -18.9 -18.1 -19.3 27<br />
3-9012 <strong>20</strong>.2 1.9-7.0 -<strong>20</strong>.6 -16.8 -22.6 12<br />
*n- number of analyzed samples from indicated boreholes<br />
159
Tarasov, AM, et, al.: Isotopic composition of the massive ground ice..,<br />
<strong>20</strong> I<br />
A<br />
2a 6 6a 8 12 I 2o<br />
Figure 1. Profiles of geological and ground ice conditions.<br />
A - study sites 1-90, profile 2; B - study sites 2-89, profile 1; C - study sites 2-89, profile 2. 7 - massive ice; 2 - clay, loam; 3 - silty sand;<br />
4 - loamy sand; 5 - hollows in massive ice: 6 - borehole location and number; 7 - lakes.<br />
CONCLUSIONS<br />
Massive ground ice is widespread in the Bovanenkovo<br />
geotectonic structure (Yamal Peninsula).<br />
In all studied deposits the bottom of the ice beds is flat,<br />
sub-horizontal and lie beneath the modern sea level. The<br />
ice top is rolling due to diagenetic processes (melting,<br />
thermodenudation and thermoerosion).<br />
The concentration of oxygen-18 and heavy hydrogen<br />
isotopes in the studied massive ice bodies reveals evidence<br />
of their genetic relation. The obtained results<br />
suggest that the ice had been formed below ground during<br />
the freezing of a water-bearing sandy horizon fed by atmospheric<br />
water precipitation.<br />
Tarasov, A.M. 1993. The isotopic composition of the massive ice of<br />
the Bovanenkovo gas-condensate field. In 'The Isotope in the Hvdrosphere'<br />
Abstr. if the 4th intern. Symp., Pjatigorsk, 18-21 May<br />
1993. Moscow: RAS (in Russian).<br />
ACKNOWLEDGEMENTS<br />
This work was supported by INTAS (grant 01 - 2329) and<br />
RFBR project - 02-05-64263.<br />
REFERENCES<br />
Dubikov, G.I. <strong>20</strong>02. Composition and cryogenic structure of permafrost<br />
in West Siberia. Moscow: GEOS Publisher (in Russian).<br />
Kurfurst, P.J., Melnikov, E.S., Tarasov, A.M. & Chervova, E.I. 1993.<br />
Co-operative Russian-Canadian engineering Geology Investigations<br />
of Permafrost on the Yamal Peninsula, Western Siberia. In<br />
Permafrost. Proc. of the 6th intern. Conf. on permafrost, Beijing, 5-<br />
9 <strong>July</strong> 1993, v.1: 356-361.<br />
Streletskaya, I.D. & Leibman, M.O. <strong>20</strong>02. Cryogeochemical interrelation<br />
of massive ground ice, cryopegs, and enclosing deposits of<br />
Central Yamal. Kriosfera Zemli 6(3): 15-24 (in Russian).<br />
Tarasov, A.M. 1990. Experience of applying oxygen-isotope method<br />
to study ground ice at engineering-geological survey. In E.S. Melnikov<br />
(ed.), Methods of engineering-geological survey: 118-133.<br />
Moscow: VSEGINGEO (in Russian).<br />
160
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Mapping of permafrost and the periglacial environments, Cordon del Plata,<br />
Argentina<br />
D. Trornbofto<br />
Affiliation IJnidad de Geocriologia, lnstituto Argentino de Nivologia, Glaciologia y Ciencias Ambientales [IANIGM) - CONICET,<br />
Mendoza - Argentina<br />
Regional mapping of periglacial areas and permafrost in<br />
the Cordillera Frontal, Mendoza, Argentina is unprecedented<br />
in South America - only certain particular sites or<br />
valleys have been mapped. A regional approach was<br />
made, with a focus on analyzing the surrounding limited<br />
periglacial and glaciological environments, and preparation<br />
of an inventory of glacial forms. The resulting map is<br />
intented to form the basis of the geocryological map of<br />
South America.<br />
The study area, Cordon del Plata, lies in the province<br />
of Mendoza, in the Argentine Central Andes between<br />
32”40‘ and 33”24’S and 69”45’ and 69”12’W. Geologically<br />
the mountain ranges belong to the Cordillera Frontal.<br />
The study area ranges from a height of <strong>20</strong>00 m a.s.1. to a<br />
maximum height of 6310 m, which corresponds to the<br />
peak of the Cerro El Plata. The minimum height was considered<br />
on the basis of geographical and geological limits<br />
of the area and its independence as a topographical unit.<br />
Three outstanding ranges may be distinguished however:<br />
“La Jaula” in the west, the “Cordon del Plata” in the central<br />
and northern part on the eastern border, and “Santa<br />
Clara” in the southwest. The area comprises ca. 2830<br />
km*.<br />
The MAAT of the meteorological stations of Aguaditas<br />
at a height of 22<strong>25</strong> m (1972-1983) and Vallecitos<br />
(1976-1985) at <strong>25</strong>00 m at its eastern flank are 7.7”C and<br />
6”C, respectively. The meteorological station Balcon I at<br />
3560 m on the tongue of a rock glacier at Morenas Coloradas<br />
indicates a MAAT of 1.6”C and precipitation of between<br />
500 mm (April <strong>20</strong>01 - April <strong>20</strong>02), and 630 mm<br />
(1991-1993).<br />
The vegetation ranges from shrubs to the Andean<br />
tundra above 3600 m. Above this zone and on the rock<br />
glaciers vegetation is extremely scarce.<br />
The map has been initiated by scanning and digitizing<br />
topographic maps (scale 1:100.000) using the AutoCad<br />
program Map (R) Release 3. The contour intervals are set<br />
at 100 m. For georeferencing of the obtained data and for<br />
the elaboration of a data base, the program ldrisi for Windows<br />
version 2.010 was applied. As a first step the existing<br />
data were superposed and updated with landsat images.<br />
In a second step the periglacial and glacial data by<br />
Corte and Espizua (1980) obtained from aerial photographs<br />
(scale I: 50.000) between March and May 1963<br />
were translated to the basic map, and corrections applied.<br />
For the periglacial environment studied for mapping the<br />
“facies” (in sense of Corte) of structural debris and inactive<br />
facies were taken from the inventory of glaciers. In the<br />
case of the “thermokarst facies”, it was considered<br />
whether or not they had a direct relation with debris-covered<br />
glaciers. They were included if they belonged to the<br />
environment of rock glaciers (“structural debris facies”) or<br />
to debris-covered glaciers. The 1981 inventory shows that<br />
the debris-covered glaciers preceded the areas or “facies”<br />
with thermokarst, although these are the exceptions. In<br />
those cases where moraine terminals were identified, the<br />
“roots” of rock glacier are found. This allows for the separation<br />
of the two differnet environments.<br />
Permafrost was detected by direct methods, using temperature<br />
profiles in surface drillings (up to a depth of 5 m)<br />
or indirect methods such as geophysical profiling or geomorphological<br />
interpretations by applying aerial photographs<br />
and satellite images. Field visits helped to verify<br />
the interpreted results. For active rock glaciers with southwesterly<br />
orientation the permafrost table was found at<br />
3770 m at a depth of 3 m (Balcon 11).<br />
The periglacial forms and types of permafrost were<br />
mapped according to (Trombotto <strong>20</strong>00).<br />
The geomorphological processes according to the altitudinal<br />
levels may be grouped as follows:<br />
1) from <strong>20</strong>00 to 3500 m, with periglacial fossil forms,<br />
dynamics of seasonal freezing and semi-arid shaping<br />
of slopes, eventual relict permafrost;<br />
2) semi-arid periglacial level from 3500 to 4500 m, with<br />
different types of permafrost, varied patterned ground,<br />
solifluction forms, glaciers, rock glaciers and mainly<br />
detrital slopes; and<br />
3) snow-covered level, with snow patches, approximately<br />
continuous permafrost, glaciers, rock glaciers and<br />
cryoplanation surfaces (Garleff 1977, Trombotto<br />
1991).<br />
161
With the help of the Andean geomorphologists, involved<br />
and by interpretating the cryogenic or cryotic characteristics<br />
and cryogenic processes or restrictive factors, the<br />
permafrost so far detected by various authors can be<br />
classified and the preliminary typology elaborated (Trombotto<br />
<strong>20</strong>00). Six'or possibly seven major groups can be<br />
proposed according to their expected ice content, their<br />
water potential, and their genesis (Tab. 1).<br />
The glacier zones have decreased, supposedly in relation<br />
to global change. As a result of the retreat of glaciers<br />
and the thinning of level 3 glacial and snow-covered zone,<br />
the periglacial level has enlarged, and the cryogenic surfaces<br />
occupy new higher zones with moraines and glacial<br />
sediments. However, the variations in the altitudinal periglacial<br />
limit (semi-arid or semi-humid) still have to be defined<br />
more precisely. A reliable catalogue of catastrophic<br />
slope events remains to be compiled. These events have<br />
not been described for Level 2 and are important for Level<br />
1 where human settlements are located. While the area<br />
above 4500 m is likely to have permafrost occurrence<br />
everywhere, the area between 3500 and 4500 m obviously<br />
has fewer possibilities of permafrost and is limited preferentially<br />
to the rock glacier valleys.<br />
Trornbntto: D.: Mapping of permafrost and the periglacial environments<br />
REFERENCES<br />
Corte, A. and Espirlia, 1. 1981. lnventario de Glaciares de la Cuenca<br />
del Rio Mendoza. IANIGLA-CONICET. lmprenta Farras, Mendoza<br />
.<br />
Garleff, K.; 1977. Hohenstufen der argentinischen Anden in Cuyo,<br />
Patagonien und Feuerland. Gottinger Geographische Abhandungen,<br />
Heft 68.<br />
Trombotto, D. 1991, Untersuchungen zum periglazialen Formenschatz<br />
und zu periglazialen Sedimenten in der 'Lagunita del Plata',<br />
Mendoza, Argentinien. Heidelberger Geographische Arbeiten, Heft<br />
90.<br />
Trombotto, D. <strong>20</strong>00. Survey of cryogenic processes, periglacial forms<br />
and permafrost conditions in South America. Revista do lnstituto<br />
Geologico 21 : 33-55.<br />
Table 1, Permafrost types and environments (*see Trombotto <strong>20</strong>00 for source references).<br />
Dn Andes<br />
Ice Hydric Permafrost Features Restrictors Tropical Desert Central Southern<br />
Content Potential Types andAndes Andes Andes Andes<br />
Processes (17"301-31"S) 131"-35"S) (35"-55"30'S)<br />
+B tfl 1) Creeping Active rock Creeping, Active rock Chile: Rock Argentina: Argentina:<br />
+B +I glaciers, Energy glaciers glaciers active rock active rock<br />
Cold balance, 2175 mmly glaciers as glaciers<br />
Discontinuous Precipitation<br />
indicators in the<br />
sense of Barsch<br />
2) Approxi- Cold Topography Chimborato NW Argentina: Argentina: Argentina:<br />
mate - Energy (6275 m), MAAT -1/-2"C, MAAT-2/4"C, I) 44"s: <strong>20</strong>60 rn<br />
Continuous balance, Ecuador:
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
New map of seasonal ground freezing and thawing in Mongolia<br />
(scale1:I<br />
500000)<br />
D. Tumurbaatar, Ya. Jambaljav, D. Undarmaa, M. Enkhtuul and A. Batbold<br />
lnstitufe of Geography, MAS, Ulaanbaatar, Mongolia<br />
INTRODUCTION<br />
In 1971 D. Tumurbaatar compiled a map of seasonal<br />
freezing and thawing of the ground around Ulaanbaatar<br />
city at a scale of 1:<strong>25</strong>,000. In 1975 S. J. Zabolotnik<br />
created a map of seasonal freezing and thawing of ground<br />
at a scale of 1:2,500,000. In <strong>20</strong>01 D. Tumurbaatar<br />
compiled a new map of seasonal freezing and thawing of<br />
ground in Mongolia at a scale of 1 :I 500000. It is based on<br />
40 years of research on permafrost in Mongolia (Devyatkin<br />
1976, Jambaljav <strong>20</strong>01 , Kudryavtsev 1978, Tumurbaatar<br />
1973, 1993). For this map we have used data of the<br />
physical and thermophysical properties of <strong>20</strong>00 samples<br />
and climate data from 234 meteorological stations located<br />
in the territory of Mongolia.<br />
SEASONAL FREEZING AND THAWING<br />
The following methods were used: (1) field survey and<br />
laboratory test analyses, (2) statistical analyses of field<br />
surveys and laboratory tests, (3) B.A.Kudryavtsev's equation<br />
for the determination of freezing and thawing depths,<br />
(4) The winsurfur software drawing the contours of mean<br />
annual surface temperature (MAST) and mean annual<br />
surface amplitudes of temperature (MASAT).<br />
The map of Quartenary geological deposits of Mongolia,<br />
compiled by E.V.Devyatkin, was taken as a base map.<br />
Depths of seasonal freezing and thawing of ground in<br />
Mongolia vary separately as follows:<br />
1) Depth of seasonal thawing for sandy loam and loam<br />
filling debris of glacial deposits is 2.1 - 3.7 meters, where<br />
MAST is 0 - -2"C, MASAT is 21 - <strong>25</strong>°C and ground moisture<br />
is 14 - 29.1%. Where the MAST is 0 - 4"C, the depth<br />
of seasonal freezing and thawing is 1.7 meter in the south<br />
and 3.8 meter in the north of the country.<br />
2) Depth af seasonal thawing in alluvial deposits is 2.4<br />
meter for loam filling debris and 4.7 meter for sand filling<br />
debris, where MAST is -2- O"C, MASAT is 21 - 31°C and<br />
ground moisture is 6 - 22%. Where MAST is 0 - 4"C, MA-<br />
SAT is 21 - <strong>25</strong>°C and ground moisture is 6 - 22%, the<br />
depth of seasonal freezing and thawing is 1.9 meter for<br />
sand filling debris in the south and 4.9 meter in the north.<br />
Where MAST is 4 - 7"C, depth of seasonal freezing is I .7<br />
meter in the central parts and 3.5 meter in the south of the<br />
countrv.<br />
3) Depth of seasonal thawing in lacustrine deposits is<br />
2.1 meter in the north, 3.6 meter in the central parts for<br />
clay and loam, where MAST is 0 - -2'C, MASAT is 23 -<br />
<strong>25</strong>°C and the ground moisture is 14 - 29.1%. Where<br />
MAST is 0 - 4"C, MASAT is 21 - 31°C and ground moisture<br />
is 14 - 29%, the depth of seasonal freezing and<br />
thawing is 1.7 meter in the south and 3.7 meter in the<br />
north. Where MAST is 4 - 9"C, MASAT is 21 - <strong>25</strong>"C, the<br />
depth of seasonal freezing is 1.0 meter in the south and<br />
2.8 meter in the central parts of the country.<br />
4) Depth of seasonal thawing in glacial-lacustrine deposits<br />
is 2.3 - 3.7 meter for sandy loam and loam, where<br />
MAST is 0 - -2"C, MASAT is 21 - <strong>25</strong>°C and ground moisture<br />
is 14 - 29.1%. Where MAST is 1 - 4"C, the depth of<br />
seasonal freezing is 1.9 meter in the south and 3.6 in the<br />
north of the country.<br />
5) Depth of seasonal thawing in alluvial-lacustrine deposits<br />
is 2.7 - 3.7 meter for sand and sandy loam with debris,<br />
where MAST is 0 - -1 "C, MASAT is 31 "C and ground<br />
moisture is 14 - <strong>25</strong>%. Where MAST is 0 - 4"C, MASAT is<br />
23 - 27"C, the depth of seasonal freezing and thawing is<br />
1.9 meter in the south and 3.7 meter in the north. Where<br />
MAST is 4 - 8"C, MASAT is 21 - <strong>25</strong>"C, the depth of seasonal<br />
freezing is 1.1 meter in the south and 2.8 meter in<br />
the north of the country.<br />
6) Depth of seasonal freezing in proluvial deposits is<br />
2.3 - 3.2 meter for loam and sandy loam with debris,<br />
where MAST is 1 - 4'C, MASAT is 21 - 27°C and ground<br />
moisture is 7 - 12%. Where MAST is 4 - 9"C, MASAT is<br />
21 - <strong>25</strong>"C, the depth of seasonal freezing is 1.3 meter in<br />
the south and 2.6 meter in the central parts of the country.<br />
7) Depth of seasonal freezing in proluvial-lacustrine<br />
deposits is 2.1 - 3.2 meter for loam and sandy loam with<br />
debris, where MAST is 1 - 4"C, MASAT is 21 - <strong>25</strong>°C and<br />
ground moisture is 7 - <strong>20</strong>.5%. Where MAST is 4 - 9"C, the<br />
depth of seasonal freezing is 1.2 meter in the south and<br />
2.8 meter in the central parts of the country<br />
8) Depth of seasonal freezing in eluvial deposits is 2.7<br />
- 4.1 meter for sand, where MAST is I - 4"C, MASAT is 23<br />
- 31°C and ground moisture is 8%. Where MAST is 4 -<br />
9"C, MASAT is 21 - <strong>25</strong>"C, the depth of seasonal freezing<br />
is I .9 - 3.23 meter.<br />
163
9) Depth of seasonal thawing in volcanic deposits is 4<br />
meter and more for basalt, where MAST is -2 - O"C, MA-<br />
SAT is 23 - <strong>25</strong>°C. Where MAST is 0 - 5"C, the depth of<br />
seasonal freezing is less than 4.0 meter.<br />
10) Depth of seasonal freezing and thawing in eluvial<br />
deposits is 1.9 - 4.7 meter for debris with loam and sandy<br />
loam fill, where MAST is 0 - 4"C, MASAT is 21 - 29°C and<br />
ground moisture is 7 - 22%. Where MAST is 4 - 9"C, MA-<br />
SAT is 21 - 23"C, the depth of seasonal freezing is 1.1<br />
meter in the south and 3.5 meter in the central parts and<br />
north of the country.<br />
I I) Depth of seasonal thawing in eluvial-solifluction<br />
deposits is 2.7 - 3.4 meter for loam and sandy loam with<br />
debris, where MAST is 0 - -IoC, MASAT is 23 - <strong>25</strong>°C and<br />
ground moisture is 7 - 22%. Where MAST is 4 - 9"C, MA-<br />
SAT is 21 - <strong>25</strong>"C, the depth of seasonal freezing is 2.3 -<br />
3.4 meter.<br />
12) Depth of seasonal thawing in elluvial-deluvial deposits<br />
is 3.0 - 4.6 meter for debris with sand fill, where<br />
MAST is 0 - -2"C, MASAT is 23 - <strong>25</strong>°C and ground moisture<br />
is 7 - 22%. Where MAST is 0 - 4"C, MASAT is 21 -<br />
29"C, the depth of seasonal freezing is 1.9 meter in the<br />
south and 4.6 meter in the central parts. Where MAST is 4<br />
- 9"C, the depth of seasonal freezing is I ,I in the south<br />
and 4.3 meter in the north of the country.<br />
13) Depth of seasonal thawing in delluvial-solifluction<br />
deposits is 3.0 - 4.5 meter for loam and sandy loam with<br />
debris, where MAST is 0 - -2"C, MASAT is 23 - <strong>25</strong>°C and<br />
ground moisture is 7 - <strong>20</strong>%. Where MAST is 0 - 4"C, MA-<br />
SAT is 23 - 29"C, the depth of seasonal freezing and<br />
thawing is 1.9 meter in the south and 4.4 meter in the<br />
north. Where MAST is 4 - 9"C, MASAT is 21 - <strong>25</strong>"C, the<br />
depth of seasonal freezing is 1.1 meter in the south and<br />
3.4 meter in the central parts.<br />
14) Depth of seasonal thawing in bedrock deposits is<br />
more than 4 meter for granites, where MAST is -2 - 0°C.<br />
Where MAST is 4 - 9"C, the depth of seasonal freezing is<br />
less than 4 meters.<br />
Using this new map, we have divided three regions of<br />
seasonal freezing and thawing (Fig. I).<br />
Region of seasonal thawing:<br />
There is continuous and discontinuous permafrost in<br />
this region. The ground thawing begins during the first<br />
decade of May and ends in the last decade of October.<br />
Region of seasonal freezing and thawing:<br />
Isolated permafrost is characteristic of this region<br />
(2°C). The ground freezing begins in the last decade of<br />
October and ends in the last decade of April.<br />
Region of seasonal freezing:<br />
In this region, permafrost does not exist. Ground<br />
freezing begins during the first decade of November and<br />
ends in the last decade of March.<br />
Tumubaatar, D. et. al.: New map of Seasonal ground freezing ...<br />
CONCLUSION<br />
I) On the represented map are shown the MAST, MASAT,<br />
superficial deposits and ground freezing and thawing<br />
depths.<br />
2) Minimum depth of seasonal thawing is 2.1 meter for<br />
clay and maximum depth of seasonal thawing is 4.7 meter<br />
for sand. Mean depth of seasonal thawing is 2.2 meter for<br />
loam and 3.9 meter for sandy loam.<br />
3) Minimum depth of seasonal freezing is 1 .0 meter for<br />
clay and maximum depth of seasonal freezing is 4.9 meter<br />
for sand. Mean depth of seasonal freezing is I .5 meter for<br />
loam and 4.0 meter for sandy loam.<br />
4) The seasonal thawing of the ground begins in the<br />
first decade of May and ends during the first decade of<br />
October. The seasonal freezing of the ground begins in<br />
the last decade of October and finishes in the last decade<br />
of April in the north and central parts of the country. This<br />
freezing begins in the first decade of November and ends<br />
during the last decade of March in the south of the country.<br />
Laks,shp<br />
Figure 1. Schematic map of permafrost regions<br />
REFERENCES<br />
Devyatkin, E.V. 1976. Map of Quartenary geological deposits of Mongolia.<br />
Jambaljav, Ya. <strong>20</strong>01. Regression equation of maximum and minimum<br />
monthly mean ground temperature curves. In proceedings of the<br />
International Symposium on mountain and arid land permafrost;<br />
33.<br />
Kudryavtsev, V.A. 1978. General geocryology (In Russian).<br />
Tumurbaatar, D. 1993. Seasonal freezing and thawing grounds of<br />
Mongolia. In proceeding of International Conference on permafrost.<br />
Bejing, China, V01.2; 1242-1246.<br />
Tumurbaatar, D. 1973. Seasonal freezing and thawing grounds of<br />
Mongolia. (In Mongolian).<br />
164
8th International Conference on Permafrost Extended Abstracts on Currmt <strong>Research</strong> and Newly Available Information<br />
Decadal changes in permafrost, land form and land cover near Barrow,<br />
Alaska<br />
C .E. Tweedie, R. D Hollister and P. J. Webber<br />
Department of Plant Biology, Michigan State University, East Lansing, USA<br />
INTRODUCTION<br />
A significant challenge in global change science is assessing<br />
the impact and feedbacks of climate change and<br />
land-uselland-cover change on ecosystem function,<br />
namely carbon exchange across the land-atmosphere<br />
boundary. In the Arctic, terrestrial carbon balance is regulated<br />
by complex spatial and temporal interactions between<br />
land cover, climate, land form, hydrologic balance,<br />
permafrost, and stochastic events such as fire, herbivory<br />
and human disturbance. The importance of arctic regions<br />
to global energy balance and biogeochemical and hydrological<br />
cycling is well recognized, as is the documented<br />
evidence, propensity and sensitivity for these to change<br />
(sometimes non-linearly) in response to climate change<br />
(Hinzman et al. submitted). The IPCC (<strong>20</strong>01) forecast Arctic<br />
regions as continuing to undergo among the fastest<br />
rates of climatic change on Earth over the next century.<br />
Clearly, the need for understanding the interaction between<br />
permafrost, land form, hydrologic balance, land<br />
cover and the response of ecosystem function in Arctic<br />
regions to climate change and following surface disturbance<br />
has never been greater.<br />
Most investigations of the functional response of Arctic<br />
ecosystems to climate change have been derived from<br />
manipulative experiments, modelling, and paleoenvironmental<br />
investigations with few studies illustrating a<br />
thorough and multidisciplinary examination of the interaction<br />
between ecosystem properties and processes. Limited<br />
by the availability of adequate and long-term datasets,<br />
few forecasts of ecosystem function in Arctic regions have<br />
been validated by observed changes. In the absence of<br />
long-term and integrated multidisciplinary datasets, reoccupation<br />
and retrospective analysis of research sites<br />
established several decades ago offer the best means by<br />
which decadal change in ecosystem function can be assessed<br />
or inferred.<br />
This report details careful re-sampling of vegetation<br />
plots established at Barrow, Alaska during the Tundra Biome<br />
Project of the International Biological Program (IBP<br />
1969-74). Results are discussed in light of other studies in<br />
the Barrow region documenting anthropogenic disturbance<br />
(Brown et al. 1980 and references therein) and<br />
changes in biogeochemical cycling (Oechel et al. 1995,<br />
<strong>20</strong>00).<br />
METHODS<br />
Plots varied in size from a few square meter plots nested<br />
within single land cover units to a 4.3 hectare grid spanning<br />
the majority of land cover types in the Barrow region.<br />
Following relocation of plot markers between 1999 and<br />
<strong>20</strong>02, resampling was performed using the original sampling<br />
methods. Comparison of seasonal active layer progression,<br />
microtopography, vegetation composition and<br />
abundance and high resolution land cover maps several<br />
decades apart has allowed the direction and magnitude of<br />
change in permafrost, land form, hydrologic balance and<br />
land cover to be assessed.<br />
RESULTS<br />
Seasonal active layer progression and maximum end of<br />
season active layer depths recorded between <strong>20</strong>00 and<br />
<strong>20</strong>02 were on average 2 to 10 cm greater than those recorded<br />
across all plots in the early 1970s. Comparison of<br />
the range in relative elevation of the ground surface<br />
measured at marked grid points between 1973 and <strong>20</strong>00<br />
suggest that there has been an increase of up to 10-15cm<br />
between polygon trough bases and adjacent rims. At the<br />
4.3 ha landscape level, however, there has been a decrease<br />
of 30 cm in the range of relative elevation.<br />
Change in vegetation was greatest in plots established<br />
in formerly disturbed or early successional communities<br />
situated on alluvial substrates. In undisturbed communities<br />
on non-alluvial substrates the greatest change were observed<br />
in wet and moist tundra. Dry tundra displayed little<br />
change over time. At the landscape scale, wet and moist<br />
tundra classes show a reduction in aerial coverage<br />
whereas dry tundra classes show an expansion in area.<br />
Within the 4.3 ha plot, spatial heterogeneity in land cover<br />
increased due to a decrease in patch size and increase in<br />
patch number.<br />
165
DISCUSSION<br />
Tweedie, C.E. et 31.: Decadal changes in permafrost, Barrow, Alaska.<br />
REFERENCES<br />
Differences between measurements made soon after plot<br />
establishment and almost three decades later are dramatic.<br />
Elevations lack registration relative to points of<br />
known height above sea level in initial measurements,<br />
which creates difficulty in assessing whether subsidence<br />
or soil development (accumulation of organic material) can<br />
be attributed to the decrease in the range of relative elevations<br />
recorded at the landscape level. Nonetheless, the<br />
minimal accumulation of organic matter around marker<br />
stakes in the most productive wet tundra land cover<br />
classes suggest that the observed change is due to subsidence<br />
of polygonized tundra in the gradient studied. The<br />
increase in the range of relative elevation between polygon<br />
troughs and adjacent rims suggests thermokarsting of<br />
ice wedges underlying polygon troughs. Change in vegetation<br />
and land cover also supports this hypothesis.<br />
Similar patterns of thermokarst development have<br />
been observed in nearby coastal zones where erosion of<br />
polygonized tundra is occurring. These elevation changes<br />
are potentially significant considering that most of the<br />
study area and nearby settlement of Barrow are ca. 3<br />
meters above sea level, indicating up to a 40% decrease<br />
in elevation relative to sea level and increased susceptibility<br />
to storm surges and flooding.<br />
Within the same study area, Oechel et al. (1995) have<br />
documented a change from tundra being a net sink for<br />
C02 to the atmosphere in the early 1970s to being a net<br />
source by the early 199Os, implying a positive feedback<br />
response to climate change. The increased active layer<br />
thickness and change in land cover documented in this<br />
study and the well-studied relationship of CO2 sink activity<br />
declining in response to increasing water deficit provides<br />
evidence of mechanisms driving patterns documented by<br />
Oechel et al. (1995). Although rates of water deficits appear<br />
to have increased during the snow-free period in the<br />
region over the past few decades (Oechel et at. 1995<br />
<strong>20</strong>00) and cannot be discounted, we suggest that the<br />
above results can be most strongly attributed to a shift in<br />
hydrologic balance caused by anthropogenic disturbance<br />
and altered land use patterns. These have been dramatic<br />
near Barrow since the mid 1940’s.<br />
Brown, J., Miller, P.C., Tieszen, L.L. and Bunnell, F.L. (eds). 1980. An<br />
Arctic Ecosystem: The Coastal Tundra at Barrow, Alaska.<br />
Stroudsburg: Dowden, Hutchinson & Ross Inc.<br />
Hinzman, L. et al. (29 other authors submitted <strong>20</strong>03). Evidence and<br />
implications of recent climate change in terrestrial regions of the<br />
Arctic. Climatic Change.<br />
IPCC. <strong>20</strong>01. Climate Change <strong>20</strong>01. The Scientific Basis. Contribution<br />
of Working Group I to the Third Assessment Report of the Intergovernmental<br />
Panel on Climate Change. In Houghton, J.T., Ding,<br />
Y., Griggs, D.J., Noguer, M., van der Linden, P.J., Dai, X.,<br />
Maskell, K. and Johnson, C.A. (eds.). Cambridge: Cambridge<br />
University Press.<br />
Oechel, W.C., Hastings, S.J., Vourlitis, G.L., Jenkins, M.A., Riechers<br />
G.H. and Grulke, N.E. 1993. Recent change of Arctic tundra ecosystems<br />
from a net carbon dioxide sink to a source. Nature 361:<br />
5<strong>20</strong>-523.<br />
Oechel, W.C., Vourlitis, G.L., Hastings, S.J., Zulueta, R.C., Hinzman,<br />
L. and Kane, D. <strong>20</strong>00. Acclimation of ecosystem C02 exchange in<br />
the Alaskan Arctic in response to climate warming. Nature 406:<br />
978-981.<br />
CONCLUSION<br />
This study has illustrated significant changes in ecosystem<br />
properties over the last thirty years near Barrow, Alaska<br />
and has linked these changes to documented changes in<br />
ecosystem function in the region. Ongoing research is investigating<br />
the relative importance of climate change, successional<br />
change (thaw lake cycle) and anthropogenic<br />
disturbance on ecosystem function in the Barrow area. It<br />
is proposed that the former IBP Tundra Biome Project<br />
study area and the adjacent Barrow Environmental Observatory<br />
(BEO) become key sites in the developing Circum-Arctic<br />
Environmental Observatories Network<br />
(CEON).<br />
166
8th International Conference on Permafrost Extended Abstracts an Current <strong>Research</strong> and Newly Available Information<br />
Contemporary geotechnologies providing safe operation of railway<br />
embankments in-permafrost conditions<br />
V. Ulitsky, V. Paramonov, S. Kudryavtsev ,K. Shashkin and *M. Lisyuk<br />
St. Petersburg State University of Highways: Moskovsky Pr. 8, Saint Petersburg, 190031, Russia<br />
*Georeconstrucfion Engineering Co ,Izmaylovsky Pr. 4, Saint Petersburg, 198005, Russia<br />
Frost heave and thaw related deformation of ground on<br />
Russian railways in permafrost conditions and under the<br />
influence of deep seasonal frost penetration attract plenty<br />
of attention as of the earliest moments of construction and<br />
operation of railway subgrades. In order to eradicate actual<br />
deformations and stabilise subgrade soils in heave<br />
sensitive areas foam polystyrene panels are currently employed<br />
(according to the following pattern: effectiveness<br />
equals costs plus technological effectiveness plus durability).<br />
This method of embankment reconstruction is currently<br />
construed to be of prime efficacy in potentially severe<br />
conditions. Design of such protective measures<br />
requires assessment both of thermophysical characteristics<br />
and of stressed-strained ground conditions during<br />
freeze-thaw process.<br />
To solve problems of the above-designated type we<br />
developed a numerical modelling methodology of such<br />
problems incorporated into FEM models software created<br />
by geotechnical engineers of St. Petersburg (Ulitsky<br />
<strong>20</strong>00). It allows solution of freeze-thaw problems by<br />
means of a thermal conductivity equation with consideration<br />
for phase transformations in below zero temperatures<br />
range for unsteady heat conditions in 3-D ground medium.<br />
In frozen rock phased water (or liquid solution) content<br />
varies within a whole range of negative temperatures. This<br />
variation occurs under the principle of the dynamic equilibrium<br />
state first described by N.A. Tsytovich (Tsytovich<br />
1979). In accordance to this principle the amount of unfrozen<br />
water for a certain type of non-saline ground is established<br />
based on sub-zero temperature values. As was<br />
proved through experiments, a humidity increase in a<br />
ground sample considerably higher than the humidity of<br />
top plasticity border in frozen state leads to unfrozen water<br />
content increase on account of infinitesimal water films<br />
found on ice crystals formed during freezing.<br />
Our FEM-models software features a function of liquid<br />
water content in the ground to assess latent heat of<br />
phased transitions being absorbed or released by ground<br />
depending on ground water phase fluctuations in belowzero<br />
temperatures. Migratory flow directed towards the<br />
freeze front at varying distance between the latter and the<br />
ground water level (GWL) is established based on ex-<br />
perimental data processing. Ground water level location<br />
and the average trend thereof in wintertime are assumed<br />
based on reference hydrometric materials specific for the<br />
given area (GWL - maximum for a monitoring period;<br />
trend - minimum for same period).<br />
The stress-strain state is assessed based on temperature<br />
fields allowing seasonally dependent definition of<br />
deformations and forces in subgrade structures and facilitating<br />
the anticipation of procedures targeted at the<br />
elimination of heave related adversary effects on the superficial<br />
elements of railway tracks. Currently it is possible<br />
to establish ground freeze-thaw development as well as to<br />
define deformation values of heave related uplift and thaw<br />
related settlement of any structures encountered within<br />
the ground bulk.<br />
We performed numerical modelling of freeze-heavethaw<br />
sequence on railway embankments of different<br />
heights by generalised cross-section profiles established<br />
over a period of long-term operation of railways in permafrost<br />
soils of Russian Transbaikalia and the Far East.<br />
For the studied Transbaikalia section of the Russian<br />
railways (station Mogocha, average annual subgrade<br />
temperature being in the order of 0-1 .Oe) foam polystyrene<br />
insulation allowed considerable temperature decrease on<br />
top of the embankment in winter time, however, with no<br />
guaranteed heave elimination. Embankment calculations<br />
for different polystyrene application techniques showed<br />
that the most effective application was that of 0.12 m and<br />
0.18 m on the top section of subgrade and the bank respectively.<br />
Underneath the insulation panels in the embankment<br />
body freezing temperature changed from -2.9 'C to -6.9'C<br />
(Fig. I), thus remaining within the heave temperature<br />
range. In warmer periods the applied insulation precludes<br />
thaw on top of subgrade thus ruling out embankment deformation<br />
in spring-summer time and positively conditioning<br />
stressed-strained subgrade structure (Fig. 2).<br />
To ensure safe operation of embankments in permafrost<br />
conditions with heaving ground affected by seasonal<br />
frost penetration it would be expedient to render geotechnical<br />
forecast by means of numerical modelling. Based on<br />
such forecast one could advise effective geotechnologies<br />
167
to apply in subgrade ground stabilisation both for seasonal<br />
frost penetration and thaw.<br />
Ulitsky, V. el al.: Contemporary gc otechnolagies providing safe operation ...<br />
REFERENCES<br />
Ulitsky, V.M., Shashkin, A.G., Shashkin, K.G. and Paramonov V.N.<br />
<strong>20</strong>00. FEM models software for modelling and solution of problems<br />
in construction and reconstruction by FE<br />
method.//Reconstruction of cities and geotechnical engineering/<br />
Saint Petersburg. 2 (in Russian).<br />
Ulitsky, V.M., Paramonov, V.N., Sakharov, 1.1.) Kudryavtsev, SA.. and<br />
Bakenov H.Z. <strong>20</strong>02. Calculation of frozen foundations footings on<br />
saline heave sensitive soils with thermal insulation. <strong>Proceedings</strong><br />
of the international coastal geotechnical engineering in practice.<br />
21-23 May <strong>20</strong>02. Atyray, Kazakhstan:l45-147.<br />
Tsytovich, N.A. and Chronic, J.A. 1979. interrelationship of the principal<br />
physicomechanical and thermophysical properties of coarse<br />
grained frozen soil. Bochum, 1978. - Eng. Geol. 13: 163-167.<br />
Figure 2. Epures of temperature distribution in 2 m high embankment<br />
and subsoil thereof with applied insulation in permafrost climate conditions<br />
of station Mogocha in Russian Transbaikalia for the month of<br />
September: 1 - extruded foam polystyrene; 2 - thawed ground;<br />
3 -frozen ground.<br />
168
li,,<br />
8th International Conference un Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
A new approach for interpreting active-layer observations in the context of<br />
climate change: a West Siberian example<br />
A. A. Vasiliev, N.G. Moskalenko and "J. Brown<br />
Earth Cryosphere Institute, Russian Academy of Sciences, Moscow, 119991 Russia<br />
* lnfernational Permafrost Association, Woods Hole, MA 0<strong>25</strong>43 USA<br />
The Circumpolar Active Layer Monitoring (CALM) program<br />
provides spatial and temporal observations of seasonal<br />
thaw depths at more than 100 sites located in different<br />
bioclimatic and landscape conditions of the polar regions<br />
(Brown et al. <strong>20</strong>00). Many of the sites consist of 100m to<br />
1000 m grids with considerable terrain variation within<br />
each grid. Thus the interpretation of these data is complicated<br />
because of the diversity of bioclimatic and landscape<br />
conditions found within and between monitoring<br />
sites. Different landscapes may demonstrate different responses<br />
to past, current and future climate changes.<br />
Our analysis, using data from two West Siberian sites, is<br />
aimed at developing a new approach to the interpretation<br />
of active layer monitoring data. This current report was<br />
stimulated in discussions during the CALM synthesis<br />
workshop held in November <strong>20</strong>02 at the University of<br />
Delaware and at recent annual conferences on Earth<br />
Cryosphere held in Pushchino, Russia.<br />
There are three CALM grids in West Siberia: Marre-<br />
Sale (R3,69'43' N, 66'52' E); Vaskiny Dachi (R5, 70"17'<br />
N, 6834' E); and Nadym (Rl, 65'<strong>20</strong>' N, 72'55' E). The<br />
Marre-Sale and Vaskiny Dachi sites are located in typical<br />
tundra; the Nadym site is located in northern taiga. For<br />
this report we use data from the Marre-Sale and Nadym<br />
sites only, because there are no long-term weather records<br />
for the Vaskiny Dachi site. In addition, we use data<br />
from previously established auxiliary sites located in the<br />
vicinities of CALM grids and having much longer observational<br />
records. Active layer depths are considered within<br />
the dominant landscapes of the tundra zone (R3) and the<br />
taiga zone (RI). For Marre-Sale these include peatland,<br />
mire, and five sites within the 1000m CALM grid: dry tundra,<br />
ravine, mire, wet tundra, and sand field as described<br />
by Melnikov et al. (1983). V. Dubrovin and N. Moskalenko<br />
established several of these sites prior to the CALM program<br />
(Vasiliev et al. 1998; Moskalenko et al. <strong>20</strong>01).<br />
Figure I illustrates the dependence of the active layer<br />
depth in these landscapes on the sum of positive air temperatures<br />
(DDT, degree-days T> O'C) at the Marre-Sale<br />
and Nadym sites. As a first approximation, the linear dependence<br />
is used. The slope of the straight line approximation<br />
is different for each landscape. We propose that<br />
this slope characterizes the active layer response of a<br />
particular landscape to climate variation. The maximum<br />
inclination is typical of mires (sites 2 and 5); whereas the<br />
minimum inclination is typical of flat peatlands (site I). All<br />
other landscapes are characterized by intermediate values<br />
of active layer depth versus DDT. As the absolute values<br />
of active layer depth are different, it is reasonable to use<br />
relative values for comparing responses of different landscapes<br />
to climate change with the use of a unified scale.<br />
zoo0<br />
DDT,, Marrc-Salc DDT,, Nadym DDT<br />
O "-I,; ~<br />
Figure 1. Correlation between the active layer depth (ALD), and DDT<br />
for the Marre-Sale and Nadym sites (1- peatland, 2- mire, 3-dry tundra,<br />
4- ravine, 5- mire, 6- wet tundra, 7- sand field. Sites 3- 7 are Iocated<br />
on the CALM grid).<br />
The mean value of DDT calculated from long-term air<br />
temperature records can be used as a climatic standard<br />
for a particular site (DDTst). For the Marre-Sale weather<br />
station, DDT,t equals 566C degree days (1914-<strong>20</strong>02) and<br />
for the Nadym weather station it reaches 1352C (1965-<br />
<strong>20</strong>02). In Figure I the vertical lines for DDTst represent<br />
this long-term average. A standard active layer depth<br />
(ALDst) for a particular landscape is represented at the in-<br />
169
tersections of this perpendicular with the corresponding<br />
approximation curves. These values are shown along the<br />
ordinate axis (ALDlst to ALD7$t) for Marre-Sale on the left,<br />
for Nadym on the right. These are very significant values<br />
and can be considered theoretical values of mean active<br />
layer depths calculated for particular landscapes. It should<br />
be pointed out that these values can be obtained on the<br />
basis of relatively short-term observation records (5-10<br />
years). These are the mean active layer depth (ALDst and<br />
average active layer depth for observation time (ALL,) in<br />
Table I.<br />
Table 1. Mean active layer depth (ALDst) and average active layer<br />
depth (ALDa,) for the observation period. Duration of observations enclosed<br />
in brackets.<br />
Vasiliev, A. A. et al.: Active-layer observations, climate change, West Siberian..<br />
(3) The diagram showing the dependence of the active-<br />
layer depth on the annual sum of positive air temperatures<br />
in relative units can be used for assessing current<br />
tendencies of climate variation or changes.<br />
If the values calculated on the basis of actual observational<br />
data are generally higher than l.Oj a tendency for<br />
climate warming is observed; the reverse is true when calculated<br />
values of the relative active layer depth are lower<br />
than 1.0. This diagram can also be used for estimating<br />
changes in the active layer depth under specific scenarios<br />
of climate change.<br />
*Landscapes are given in Figure 1 captions<br />
On the basis of these data, we can recalculate all<br />
available data on the ALD and DDT into relative units using<br />
corresponding standard values, and plot another diagram<br />
showing the dependence of the relative active layer<br />
depth on the relative sum of positive summer temperatures<br />
(Fig. 2). Calculations have been made for contrasting<br />
landscapes (mires and peatlands). The resulting<br />
curves for the other landscapes should occupy intermediate<br />
positions. This diagram makes it possible to assess<br />
current tendencies in the active layer development (Fig.<br />
2). If most of the observational data tends to be concentrated<br />
in the right part of the diagram, then this indicates<br />
climate warming and an increase in the active layer depth.<br />
By contrast, the concentration of data in the left part of the<br />
diagram attests to a cooling trend with a corresponding<br />
decrease in the active layer depth.<br />
Similar calculations have been made for the Nadym<br />
monitoring site (northern taiga zone). It is interesting that<br />
the resulting diagrams for mires and peatlands have the<br />
same shape. The response of these landscapes to the<br />
degree of summer warming (DDT values) does not depend<br />
on the climatic zone (Melnikov at al., this volume).<br />
The following conclusions can be stated from this<br />
analysis:<br />
(I) The suggested method makes it possible to calculate<br />
mean maximum depths of the active layer for longterm<br />
periods on the basis of relatively short-term observational<br />
records.<br />
(2) The response (as manifested by the active layer<br />
depth) of seven dominant types of landscapes of the typical<br />
tundra zone and two dominant types of landscapes of<br />
the northern taiga zone to climate changes is evaluated<br />
quantitatively. The active layer depth in mires is very sensitive<br />
to climate changes; the least sensitive is typical of<br />
peatlands.<br />
" -<br />
Coolilg<br />
Warmk g<br />
.. . . ".<br />
0,s 0,7 09 I,l 1,3 1,s<br />
DDT rel.<br />
Max.T, mire - Max NT, mire<br />
Mi. T, peatland - - Min NT, pealand<br />
Figure 2. Diagram of the active layer response to warming and cooling<br />
in tundra (T) and northern taiga (NT) of West Siberia.<br />
REFERENCES<br />
Brown, J., Hinkel K.M., and Nelson, F.E. <strong>20</strong>01. The Circum-polar active<br />
layer monitoring (CALM) program: research designs and initial<br />
results. Polar Geography 24(3):165-<strong>25</strong>8.<br />
Melnikov, E.S. (ed.) 1983. Landscapes of a permafrost zone in the<br />
West Siberian gas province. Novosibirsk: Nauka.<br />
Melnikov, E.S., Leibman, M.O., Moskalenko, N.G., and Vasiliev A.A.<br />
<strong>20</strong>03. Comparative analysis of active layer monitoring at the<br />
CALM spatial variability at European Russian Circumpolar Active<br />
Layer (CALM) sites (this volume).<br />
Vasiliev, A.A., Korostelev Yu.V., Moskalenko N.G., and Dubrovin V.A.<br />
1998. Active layer monitoring in West Siberia under the CALM<br />
program (database). Earth Cryosphere ll(3): 87-90.<br />
Moskalenko, N.G., Korostelev, Yu.V., and Chervova, E.I. <strong>20</strong>01. Active<br />
layer monitoring in northern taiga of West Siberia. Earth<br />
Cryosphere V(1): 71-79.<br />
170
Influence of seasonal active-layer thickness on broad-scale vegetation<br />
dynamics in the permafrost zone<br />
S. Venevsky<br />
Obukhov Institute for Atmospheric Physics, Moscow, Russia and Potsdam Institute for Climate Impact <strong>Research</strong>, Potsdam,<br />
Germany<br />
SPECIFIC BIOLOGICAL AND PHYSICAL<br />
PROCESSESIN THE PERMAFROSTZONE<br />
Change of active-layer thickness<br />
Continuous and discontinuous permafrost is the factor most<br />
unique for the tundra and taiga environment. Dynamics of the<br />
permafrost table, or active-layer thickness depend on the local<br />
surface energy balance, reflecting above all air temperature<br />
as modified by soil thermophysical properties, vegetation,<br />
snow and solar radiation. In particular, the presence of a<br />
thick forest floor and organic peat layer significantly lowers<br />
soil temperatureand active-layer thickness, as low thermal<br />
conductivity of the organic matter effectively insulates the<br />
mineral soil. A snow cover in autumn limits frost penetration,<br />
while in spring it functions as an insulating blanket to delay<br />
thawing of soil active-layer. Active-layer thickness and low<br />
soil temperature provides a strong feedback on vegetation at<br />
high latitudes. Low soil temperatures influence forest prcduG<br />
tivity by restricting available nutrients due to low dewmposition<br />
rates and by reducing water uptake by trees due to an<br />
increase in root resistance and the viscosity of water. Trees<br />
in spring and the first half of summer can also be subject b<br />
severe water stress by high evaporation demands when their<br />
roots are still frozen and cannot provide therequiredwater<br />
uptake.<br />
SIMULATION OF VEGETATION DYNAMICS IN THE<br />
PERAMFROST ZONE<br />
Ecological objective of the study<br />
Annual and seasonal variation of active-layer thickness affects<br />
local soil moisture status and, thus regulates the water<br />
available to plants and fuel moisture in fire-prone vegetation<br />
communities, especially in spring. One should investigate the<br />
role played by seasonal and annual variation of active-layer<br />
thickness in regional fire regimes, vegetation composition and<br />
carbon stocks on a broad scale in the permafrost zone.<br />
Basic vegetation model<br />
The Lund-Potsdam-Jena Dynamic General Vegetation Model<br />
(LPJ-DGVM) (Sitch et al. <strong>20</strong>03) combines mechanistic<br />
treatments of terrestrial vegetation dynamics, carbon and<br />
water cycling in a modular framework. Key features include<br />
fast feedbacks between photosynthesis and water balance<br />
routines and representations of slower ecosystem-level processes<br />
including vegetation resource competition, growth, establishment<br />
and mortality, soil and litter carbon turnover. To<br />
link the fire regime and its effects on vegetation dynamics, a<br />
general fire model was incorporated into the LPJ-DGVM.<br />
This model estimates fire conditions based on soil moisture.<br />
Forest fires<br />
After climate and soil, fire is the most important natural factor<br />
influencing nutrient cycling, chemical and physical processes<br />
Principal method of modification of the Vegetation model for<br />
the permafrost zone<br />
In order to meet the objectives of the study, the LPJ-DGVM<br />
in soil and forest dynamics (regeneration, growth, species was revised and considerably modified for the permafrost<br />
composition and mortality) in a boreal part of the permafrost zone by including the active-layer thickness as one of the<br />
zone. Depending on flame length and severity of fire, vegeta- key model variables (Venevsky <strong>20</strong>01). Climate variables<br />
tion and topography characteristics, and presence or absence (daily temperature, precipitation and incoming radiation) define<br />
of permafrost, the influence of fire on forest ecosystem is active-layer thickness, soil moisture status and vegetation<br />
highly variable. Frequent fires can radically change the local status. A daily value of the active layer thickness affects<br />
biogeochemical status, redistribute and mobilize nutrients and water uptake by roots, runoff and percolation of water in the<br />
remove the forest floor layer. Therefore, fire in the boreal zone soil and, therefore, vegetation production and soil moisture<br />
can be considered as a natural rejuvenating force, which content. The daily soil moisture content determines the prob<br />
rapidly returns climax forests to the initial stage of their devel- ability of fire, vegetation production status and determines<br />
opment cycle.<br />
thermophysical properties of soil (and as a result, a current<br />
active-layer thickness). Photosynthesis and water uptake by<br />
171
plants are joint processes which significantly influence daily<br />
soil moisture content. Vegetation can be removed by fire in<br />
the model and affects fire by an amount of accumulated litter<br />
(fuel), The dynamics of the permafrost table, or active-layer<br />
thickness, depends on the local surface energy balance, reflecting<br />
first of all air temperature as modified by soil therm@<br />
physical properties, vegetation, snow and solar radiation<br />
(Kudryavtsev et al. 1974). Due to variations in snow cover<br />
or vegetation type and corresponding changes in heat transport<br />
and radiative energy exchange, surface temperatures<br />
may differfromtheairtemperature by several degrees. A<br />
simple periodic trigonometric function with the maximum annual<br />
thaw depth as an amplitude was applied to represent<br />
daily depth profile of seasonal thawing and freezing, which<br />
starts from above. Three regions with measured seasonal<br />
thawing depth were taken for validation of the active layer<br />
model, two in Siberia (Central Jakutia) and one in Alaska<br />
(Kuparuk river basin).<br />
RESULTS OF SIMULATION WITH HISTORICAL<br />
CLIMATE DATA<br />
Set of experiments<br />
To assess the influence of permafrostandfireregimes for<br />
vegetation structure and carbon poolslexchange, the modified<br />
version of the LPJ-DGVM was run in two modes, with and<br />
without active-layer processes. For the global historical<br />
simulation within the LPJ-DGVM, the CRU05 1901-1995<br />
0.5"x0.5" monthly climate data, provided by the Climate<br />
<strong>Research</strong> Unit, University of East Anglia, UK, was used.<br />
Venevsky. S.: Active-layer thickness, vegetation, dynamics ...<br />
Comparison with observations<br />
Aboveground vegetation biomass differs rather significantly in<br />
the runs with and without active-layer processes. The vegetation<br />
carbon content in the Asian part of Russia changed from<br />
7.5-12.5 kgClm2 to 2.5-10 kgClm2 and generally decreased<br />
1.5-2.5 times, when the processes seen in the permafrost<br />
zone were included in the LPJ-DGVM. The fire return intervals<br />
in western and eastem Siberia decreased two to four<br />
times, especially in the continental areas, in comparison with<br />
the unmodified model version. The average growing stock of<br />
Larch stands by the nine administrative units of Siberia,<br />
simulated by the two LPJ-DGVM versions for the year 1993,<br />
was compared with the data of Russian Forest State The<br />
area of the administrative units varies from 0.4 to 3 IO6km2.<br />
The mean square relative deviation of simulated against ob<br />
served growing stock was 26% for the modified LPJ-DGVM<br />
version, compared with 70% accuracy obtained with the<br />
original version. The growing stock for the largest northem<br />
administrative units, Magadanskaia oblast', Sakha (Jakutia)<br />
and Evenkiiskaia A0 decreased 1.5 to 2 times after the<br />
model modification and became close to reality.<br />
CONCLUSION<br />
The significant improvement of results for the vegetation carbon<br />
in the permafrost zone is mainly influenced by changes<br />
in the hydrological regime with associated transitions in the<br />
fire regimes. Therefore, interaction of fire regimes and<br />
thawlfreeze processes are very important for vegetation<br />
structures in the permafrost zone.<br />
REFERENCES<br />
Venevsky, S. <strong>20</strong>01. Broad-scale vegetation dynamics in northeastern<br />
Eurasia - observations and simulations, Universitat fur<br />
Bodenkultur, Vienna, 150 pp.<br />
Sitch, S., Smith, B., Colin Prentice I. and LPJ consortium. <strong>20</strong>03.<br />
Evaluation of ecosystem dynamics, plant geography and terrestrial<br />
carbon cycling in the LPJ Dynamic Global Vegetation<br />
Model, Global Change Biology 9: 161-185.<br />
Kudryavtsev, V.A., Garagulya, L.S., Kondrat'yeva, K.A. and Melamed,<br />
V.G. 1974. Fundamentals of frost forecasting in geological engineering<br />
investigations. Nauka, Moscow, 431 pp.<br />
172
Permafrost Monitoring in <strong>Switzerland</strong> (PERMOS) in the pilot stage<br />
D.S. Vonder Muhll, *R. Delaloye, ** W. Haeberli, **M. Hoelzle and ***E Krummenacher<br />
Delegate for Permafrost of the Glaciological Commission; Universities of Basel and <strong>Zurich</strong>, <strong>Switzerland</strong><br />
*Institute of Geography, University of Fribourg, <strong>Switzerland</strong><br />
**Glaciology and Geomorphodynamics Group, Department of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
***Geotest AG, Davos, <strong>Switzerland</strong><br />
INTRODUCTION<br />
The variation of the glacier tongues has been observed<br />
systematically in <strong>Switzerland</strong> since the 1850s. The Swiss<br />
Academy of Sciences (SAS) and the Swiss Alpine Club<br />
(SAC) were the driving institutions behind ensuring and<br />
maintaining the measurements. Permafrost is a thermal<br />
phenomenon of the subsurface, and with the snow it complements<br />
the cyrosphere. In contrast to glaciers and snow,<br />
systematic scientific investigations of Alpine permafrost<br />
started some 30 years ago.<br />
The drilling through the Murtel-Corvatsch rock glacier<br />
in 1987 triggered a number of follow-up and parallel activities<br />
in permafrost research at several Swiss institutes.<br />
Concepts for a Swiss monitoring network on permafrost<br />
were developed (Haeberli et al. 1993, Glaziologische<br />
Kommission <strong>20</strong>00) parallel to efforts initiated by the International<br />
Permafrost Association (IPA, Frozen Ground<br />
1995; Brown 1997). While the IPA monitoring strategy was<br />
developed mainly for high-latitude plain permafrost areas,<br />
PERMOS is adapted to the low-latitude mountain perrnafrost.<br />
development can be analyzed and monitoring parameters<br />
adapted if necessary<br />
During the pilot phase, PERMOS comprises IOboreholes<br />
(which all were drilled within scientific projects with<br />
mainly non-monitoring aims) and 10 BTS areas (some of<br />
which have been observed for several years already) plus<br />
about two flights per year for aerial photographs. The distribution<br />
of the sites is somewhat random (Fig. 1). Spatial<br />
gaps will be filled after the pilot phase.<br />
OBJECTIVES<br />
The main goal is to document status and long-term variations<br />
of Alpine permafrost. PERmafrost Monitoring in<br />
<strong>Switzerland</strong> (PERMOS) is based on three elements:<br />
1. recording thermal changes in permafrost, in particular<br />
permafrost temperatures in boreholes,<br />
2. determining the lateral distribution pattern near the<br />
lower permafrost boundary using the BTS (Bottom<br />
Temperature of the Snow cover in March) - method in<br />
combination with single-channel temperature-loggers,<br />
and<br />
3. taking aerial photos to build an archive for photogrammetrical<br />
determination of geomorphologic surface<br />
changes due to changes in permafrost characteristics.<br />
As a consequence, the processes involved can be investigated<br />
and modelled to improve the understanding<br />
and knowledge of Alpine permafrost. In addition, long-term<br />
Figure 1. Distribution of the PERMOS sites (drillings and BTS areas)<br />
during the pilot phase in the contours of <strong>Switzerland</strong>.<br />
MANAGEMENT SETUP<br />
The network was initiated by the Swiss Adhering Body of<br />
the IPA, and the concept (Glaziologische Kommission<br />
<strong>20</strong>00) approved by the Glaciological Commission of the<br />
Swiss Academy of Sciences (SAS). During the pilot<br />
phase, PERMOS is financed by the SAS, the Swiss Forest<br />
Agency and the Federal Office for Water and Geology.<br />
Measurements are done by eight institutions (Academia<br />
Engadina Samedan, ETH <strong>Zurich</strong>, the Universities of<br />
Basel, Bern, Fribourg, Lausanne and <strong>Zurich</strong>, SFISAR Davos).<br />
Annual reports are published by the Glaciological<br />
Commission of the SAS.<br />
173
INITIAL RESULTS<br />
Permafrost temperatures have been recorded at Murtel-<br />
Corvatsch since 1987 (Fig. 2). Processes and causes for<br />
variations are discussed by Vonder Muhll (<strong>20</strong>01). Similar<br />
temperature series are recorded at all PERMOS drill sites.<br />
Permafrost temperature at IOm depth range is between<br />
-2°C and 0°C. Thickness of the active layer is determined<br />
where a data logger records temperatures.<br />
The BTS-area of Alpage de Mille (22<strong>20</strong> to 2460 m<br />
a.s.1.) has been observed systematically since 1996.<br />
Every year at around the same date, BTS values are<br />
measured at more that 60 points to investigate variations<br />
from one year to another. While the pattern is about the<br />
same every year, BTS values vary up to 3°C from one<br />
year to the next (Vonder Muhll et al. <strong>20</strong>01). The BTS<br />
working group of PERMOS is analyzing the set up of the<br />
BTS areas and adapting the method for the second pilot<br />
phase.<br />
Due to weather conditions, no aerial photos were taken<br />
in <strong>20</strong>01. In <strong>20</strong>02, photos were taken in the Upper Engadin<br />
region (Corvatsch-Murtel).<br />
OUTLOOK<br />
by Swiss Federal Departments, together with the Swiss<br />
Glacier Network, in about three years. It is important that<br />
the measurements are taken, interpreted and coordinated<br />
by the scientific community of the Swiss Adhering Body of<br />
I PA.<br />
REFERENCES<br />
Brown J. 1997. Disturbance and recovery of permafrost terrain. In:<br />
Disturbance and recovery in Arctic lands (Ed. R. M. M. Crawford).<br />
Kluwer Academic Publishers: 167-178.<br />
Glaziologische Kommission <strong>20</strong>00. Aufbau eines Permafrost-<br />
Beobachtungsnetzes in der Schweiz - PERMOS (Konzept und<br />
Anhang). 5 plus appendix.<br />
http://www.unibas.ch/vr-forschung1PERMOS<br />
Haeberli W., Hoelzle M., Keller F., Schmid W., Vonder Muhll D. and<br />
Wagner S. 1993. Monitoring the long-term evolution of mountain<br />
permafrost in the Swiss Alps. Sixth International Conference on<br />
Permafrost, Beijing, <strong>Proceedings</strong> 1: 214-219.<br />
Vonder Muhll D. <strong>20</strong>01. Thermal variations of mountain permafrost: An<br />
example of measurements since 1987 in the Swiss Alps. In Visconti<br />
et al (eds.): Global Change in Protected Areas.(Kluwer Academic<br />
Publisher): 83-95.<br />
Vonder Muhll D., Delaloye R., Haeberli W., Holzle M and Krummenacher<br />
B. with contributions of 13 others, <strong>20</strong>01: Permafrost Monitoring<br />
<strong>Switzerland</strong> PERMOS, 1, Jahresbericht 19991<strong>20</strong>00, 32p<br />
plus appendix.<br />
http://www.unibas.ch/vr-forschunglPERMOS,<br />
PERMOS is part of the Global Terrestrial Network - Permfrost<br />
GTN-P. I t is on track to be established and financed<br />
MurWCorvattsch 2t1987<br />
I l I I I I<br />
I I I I l I<br />
I I I , I I<br />
""b-"..P"<br />
I , ,<br />
b m m O r t 4 A * m m b m m O r<br />
m m m m m m m m m m ~ m m o m 2<br />
Figure 2. Permafrost temperature at 11.6 m depth in borehole 211987 through MurteI-Corvatsch rock glacier. In the winters 198811989, 199511996<br />
and <strong>20</strong>011<strong>20</strong>02 major snowfall came only in JanuarylFebruary, causing a strong cooling of the permafrost.<br />
1 74
Methanogenesis under extreme .environmental conditions in permafrost<br />
Soils: a model for exobiological processes<br />
D. Wagner, S. Kobabe and E.". Pfeiffer<br />
Alfred Wegener Institute for Polar and Marine <strong>Research</strong>, <strong>Research</strong> Unit Potsdam, Germany<br />
INTRODUCTION<br />
Permafrost within the solar system exists on seven planets<br />
as well as several moons, which are characterized by cold<br />
environmental conditions. Of special interest to present-day<br />
research in astrobiology is the comparability of environmental<br />
and climatic conditions of the early Mars and Earth. Martian<br />
and terrestrial permafrost areas show similarmorphological<br />
structures, suggesting that their development is based 011<br />
comparable processes. New high-resolution images of the<br />
Mars Orbiter Camera (MOC) imply water liquid in form or as<br />
ground ice on Mars, which is an essential bio-physical requirement<br />
for the survival of micro-organisms in permafrost.<br />
Terrestrial permafrost, in which micro-organisms have survived<br />
for several million years (Rivkina et al. 1998), is an<br />
ideal model for comparative system studies regarding the<br />
evolution, adaptation and survival strategy of microorganisms<br />
under extreme conditions.<br />
METHANOGENESIS<br />
The microbial methane production (methanogenese) is the<br />
terminal step during the decomposition of organic matter in<br />
permafrost soils. Responsible for the methane production is a<br />
small group of highly specialized microorganisms, called<br />
methanogenic archaea, which are considered to be one of the<br />
initial organisms from the beginning of life on Earth. They are<br />
regarded as strictly anaerobic microbs, which can grow and<br />
survive only under anoxic conditions, like in wet tundra environments.<br />
Compared to Martian conditions, such a metab<br />
lism would be favourable because the atmosphere on Mars<br />
is dominated by C02 (95.3 %). Furthermore, they are characterized<br />
by litho-autotrophic growth, whereby energy is<br />
gained by the oxidation of hydrogen, and carbon dioxide can<br />
be used as the only carbon source (Deppenmeier et al.<br />
1996). Organic compounds, which were not verified an<br />
Mars, are not required for growth. The special metabolism<br />
qualities of the methane-forming archaea are the reason that<br />
they are regarded as key organisms for further investigation<br />
the adaptation strategies and long-term survival in extreme<br />
environments such as those with terrestrial permafrost (Wagner<br />
et al. <strong>20</strong>01).<br />
MICROBIAL METHANE PRODUCTION UNDER EX-<br />
TREME CONDITIONS<br />
The field investigations were carried out on the Samoylov island<br />
(N 72", E 126") located in theLena DeltalSibena. The<br />
study site represented an area of typical polygonal patterned<br />
grounds with ice wedges. The permafrost soils were classified<br />
as Glacic Aquifurbels and Typic Hisforthels with a<br />
maximum thaw depth between 30 and 55 cm. The average<br />
air temperature was -14.7 "C with a min. of -47.8 "C in<br />
January and a max. of t18.3 "C in <strong>July</strong>.<br />
The investigation of in situ methane production revealed<br />
methane formation in the boundary layer to the permafrost at<br />
temperatures between 0.6 and 1.2"C (Fig. 1) with a rate cf<br />
about 1.0 nmol CH4 h-1 g-1. Without any additional substrate,<br />
the methane production varied between 0.3 and 4.7 nmol<br />
CH4 h-1 g-1 (Fig. 2). The highest activity could be determined<br />
in the peat layer of the upper soil at an average temperature cf<br />
10 "C. Nevertheless, the comparison of different vertical pr@<br />
files to the methane production showed that the activity of the<br />
micro organisms is independent of the soil temperature. After<br />
addition of methanogenic substrates (acetate, Hz), the activity<br />
drastically increased. The methane production in the peat<br />
layer with H2 as substrate was about 2 times higher (13.0<br />
nmol CH4 h-1 9-1) compared with acetate (7.2 nmol CH4 h-1<br />
g-1) as substrate, while above the permafrost table activity the<br />
was in the same order of magnitude (approx. 1.5 nmol CH4<br />
h-1 g-1) with both substrates. Even the incubation of soil mate<br />
rial from the active layer with H2 as a substrate showed a<br />
significant methane production rate at -3 "C (0.1 - 11 -4 nmol<br />
CH4 h-I g-I) and -6 "C (0.08 - 4.3 nmol CH4 h-1 g-I). The<br />
existence of a methanogenic community, which has adapted<br />
well to the low in situ temperature of permafrost soils is assumed<br />
by our results.<br />
In accordance with Panikov (1997), psychrophilic bacteria<br />
have a significant part in the microbial community in cold<br />
environments like permafrost soils. However, the isolation<br />
175
and characterization of methanogenic archaea from permafrost<br />
soils should clarify the possible growth characteristic (psychrophile<br />
or psychrotroph) and the ecological significance d<br />
these organisms.<br />
80<br />
A<br />
CH<br />
0 <strong>20</strong> 40 60 80 100 1<strong>20</strong> 140 160 180 <strong>20</strong>0<br />
time Ihl<br />
Figure 1. In situ methane production in the boundary layer<br />
permafrost in the soil of the polygon depression.<br />
-c H,<br />
&acetate<br />
-without<br />
D 2 4 6 8 I 0 1 2 1 4<br />
Cy production rate [nrnol h'h'q<br />
to the<br />
bial population will be influenced by the natural permafrost<br />
system and by the interaction of the combined parameters.<br />
REFERENCES<br />
Deppenmeier, U., Muller, V., Gottschalk, G. 1996. Pathways of energy<br />
conservation in methanogenic archaea. Archive of Microbiology<br />
165: 149-1 63.<br />
Panikov, N.S. 1997. A kinetic approach to microbial ecology in arctic<br />
and boreal ecosystems in relation to global change, In W.C.<br />
Oechel (ed), Global Change and Arctic Terrestrial Ecosystems<br />
Berlin: Springer. 171 -188.<br />
Rivkina, E., Gilichinsky, D., Wagener, S., Tiedje, T., McGrath, J. 1998.<br />
Biogeochemical activity of anaerobic microorganisms from<br />
buried permafrost sediments. Geomicrobiology 15: 187-193.<br />
Wagner, D., Spieck, E., Bock, E., Pfeiffer, E". <strong>20</strong>01. Microbial life in<br />
terrestrial permafrost: methanogenesis and nitrification in Gelisols<br />
as potentials for exobiological processes. In G. Horneck, C.<br />
Baumstark-Khan (eds), Astrobiology: the quest for the conditions<br />
of life Berlin: Springer. 143-159.<br />
Wagner, D., Wille, C., Pfeiffer, E.". in preparation. Simulation of<br />
annual freezing-thawing cycles in a permafrost microcosm for<br />
assessing microbial methane production under extreme conditions.<br />
Permafrost and Periglacial Processes, submitted,<br />
Figure 2. Vertical profile of methane production rates without any<br />
additional substrate as well as with HP and acetate in the soil of the<br />
polygon depression.<br />
First results concerning the diversity of methanogenic microflora<br />
by fluorescence in situ hybridization (FISH) indicated<br />
that Hz-using genera like Methanobacterium and Methanomicrobiurn<br />
dominated over methylotrophic and acetotrophic<br />
methanogenic archaea. Enrichment cultures, which were<br />
dominated by the genera Methanosarcina, showed that these<br />
organisms grow not only with the preferred substrate methanol<br />
but also with C02 and HL as the only carbon and energy<br />
source.<br />
CONCLUSIONS AND FUTURE PERSPECTIVES<br />
Methanogenic archaea, which can live solely on the basis of<br />
water,minerals,carbondioxideandhydrogen, survived in<br />
terrestrial permafrost for several millions of years. Since they<br />
are still active, they had to develop different strategies to resist<br />
extreme conditions like salt stress, physical damage by<br />
ice crystals and background radiation. The presented results<br />
demonstrate an essential basis for the understanding of the<br />
origin of life. The studies can be used as a model for exobiological<br />
processes to find traces of life in extraterrestrial habitats<br />
and to look for possible protected niches e.g. on Mars.<br />
Apart from the presented studies, simulation experiments<br />
with permafrost microcosms can supply important results for<br />
the understanding of microbial life under extreme conditions<br />
(Wagner et al. <strong>20</strong>03). Simulation of the freezing-thawing cycle<br />
can help to understand the problem in which way the micro-<br />
176
A physiognomic-based circumpolar Arctic vegetation map<br />
D. A. Walker, M. K. Raynolds and, *N. G. Moskalenko<br />
Alaska Geobotany Center, lnsfifute of Arctic Biology, University of Alaska Fairbanks<br />
*Earth Ctyosphere Institute, Siberian Division, Russian Academy of Sciences, Russia<br />
The Circumpolar Arctic Vegetation Map (CAVM) depicts<br />
the distribution of <strong>20</strong> vegetation-physiognomy units in the<br />
Arctic (CAVM Team <strong>20</strong>03). The map is based on the principle<br />
that environmental controls such as climate, topography,<br />
soil chemistry and floristic provinces determine<br />
which plants grow across the Arctic. The most important<br />
environmental control of plant growth in the Arctic is summer<br />
temperature. Temperature and vegetation data were<br />
used to define bioclimatic subzones.<br />
For purposes of display, the <strong>20</strong> units of the CAVM were<br />
summarized into a much simpler map with 7 units (Table<br />
l), based on dominant growth forms. The distribution of<br />
these categories by bioclimatic subzone (Fig. 1) is shown<br />
in Table 2.<br />
Table 1. Condensed physiognomic legend with CAVM units.<br />
Glaciers: 17. Nunatak area; 18. Glacier<br />
Barren, herbs: 1. Cryptogam, cushion forb barren;<br />
2. Cryptogam barren (bedrock); 3. Noncarbonate<br />
mountain complex; 4. Carbonate<br />
mountain complex<br />
Graminoid, prostrate shrubs: 5. Rushlgrass, forb<br />
cryptogam tundra; 6. Grarninoid, prostrate dwarf<br />
shrub, forb tundra<br />
Prostrate, hemi prostrate shrubs: 9. Sedge/grass,<br />
moss wetland; 13. Prostrate dwarf shrub, herb<br />
tundra; 14. Prostratdhemiprostrate dwarf shrub<br />
tundra<br />
Graminoid, dwarf shrubs: 7. Non-tussock sedge,<br />
dwarf-shrub, moss tundra; 8. Tussock sedge,<br />
dwarf shrub moss tundra<br />
Erect dwarf shrubs, low shrubs: 10. Sedge, moss,<br />
dwarf shrub wetland; 11. Sedge, moss low shrub<br />
wetland; 15. Erect dwarf shrub tundra; 16. Low<br />
shrub tundra<br />
Water: 19. Lakes. <strong>20</strong>. Ocean<br />
Figure 1. Bioclimate subzones of the circumpolar Arctic.<br />
The glacier category excludes the Greenland Ice Sheet<br />
(1697 km*), which spans numerous bioclimatic subzones.<br />
Glaciers cover the largest area in Subzone C, but cover<br />
the greatest percentage of Subzone A. Barren vegetation<br />
types are common in Subzones A and B, especially on the<br />
calcareous soils of the Canadian Arctic Islands. Barren<br />
areas in Subzone C are mainly on the Canadian Shield<br />
and the Taymyr Mountains; and in Subzones D and E, are<br />
due to the mountains of Chukotka and Alaska.<br />
Graminoid-dominated vegetation types with prostrate<br />
shrubs are found in the southern Canadian Arctic Islands<br />
(Subzones B and C). Graminoid types with erect shrubs<br />
(including tussock tundra) are more common in Alaska<br />
and Yakutia (Subzones D and E).<br />
Prostrate-shrub dominated vegetation types are common<br />
on the southern Canadian Arctic Islands, northern<br />
mainland Canada and the Taymyr Peninsula (mainly Subzone<br />
C). Erect shrubs are found in western Russia,<br />
southern Chukotka, southern Keewatin and the Ungava<br />
Peninsula (mainly Subzone E).<br />
177
~<br />
~<br />
~ prostrate<br />
I<br />
Vegetation Physiognomy<br />
C"" 1<br />
Water and non-arctic<br />
Q Glaciers<br />
Lz] Barren, herbs<br />
" "_ I_<br />
Gramhold,<br />
shrubs<br />
Prostrate, hemiprostrate<br />
shrubs<br />
Grarninoid,<br />
dwarf shrubs<br />
Erect dwarf shrubs,<br />
low shrubs<br />
.. .<br />
Figure 2. Distribution of vegetation physiognomy in the circumpolar Arctic<br />
Table 2. Area of physiognomic categories in bioclimate subzones<br />
(1000 km2).<br />
T - Subzo -<br />
Physiognomic Unit - A - B - c -<br />
11 E Totul<br />
Glaciers<br />
B,men, herbs<br />
Graminoid, prostrate<br />
shrubs<br />
Prostrate, hemiprostrate<br />
shruhs<br />
Graminoid, dwarf<br />
shrubs<br />
Erect dwarf shrubs,<br />
95<br />
61<br />
2<br />
37<br />
0<br />
63<br />
242<br />
71<br />
130<br />
0<br />
127<br />
332<br />
375<br />
305<br />
34<br />
74<br />
343<br />
162<br />
19<br />
609<br />
2<br />
23 I<br />
0<br />
68<br />
505<br />
360<br />
1<strong>20</strong>8<br />
610<br />
560<br />
1148<br />
low shrubs<br />
0 0 - 0 - 272 - 924 1196<br />
Total<br />
194 506 1174 1479 1729 SO83<br />
---<br />
1<br />
zone D is mostly graminoid, dwarf shrubs. Subzone E is a<br />
mix of erect shrubs and graminoid, dwarf shrubs.<br />
ACKNOWLEDGEMENTS<br />
This research project is funded by National Science Foundation<br />
Grant OPP-9909929, with support from the US Fish<br />
and Wildlife Service and the Circumpolar Arctic Flora and<br />
Fauna Project.<br />
REFERENCES<br />
Subzone A is predominantly glaciers and barren. Sub- Circumpolar Arctic Vegetation Mapping Team, in press. Circumpolar<br />
zone B is predominantly barren and prostrate shrubs.<br />
Arctic vegetation. Conservation of Arctic Flora and Fauna (CAFF)<br />
Map No. 1.<br />
Subzone C is a mix of barren areas (Canadian Shield);<br />
Scale 1 :7,500,000. US Fish and Wildlife Service, Anchorage,<br />
AK.<br />
graminoid, prostrate shrubs; and prostrate shrubs. Sub-<br />
178
Permafrost and the Colville River Delta, Alaska<br />
H. J. Walker<br />
Department of Geography and Anthropology, Louisiana State University, Baton Rouge, LA 70803-4105, USA<br />
Although the study of frozen ground has a fairly long history,<br />
it was not until 1943 that Siemon Muller first used the<br />
word permafrost in a classified volume for the US. Army<br />
entitled Permafrost or Permanently Frozen Ground and<br />
Related Engineering Problems. In 1947, after declassification,<br />
this monograph was published as a hard-cover<br />
book and then became available to the public (Muller<br />
1947). It can be surmised that Professor Muller, in his<br />
Stanford University office, never anticipated that within <strong>20</strong><br />
years there would be an international conference on the<br />
subject and that an international permafrost association<br />
would soon follow.<br />
<strong>Research</strong> on the geomorphology of the Colville River<br />
delta, which was begun in 1961, included permafrost as a<br />
main item of concern. At the First International Conference<br />
on Permafrost, held at Purdue University, USA in 1963<br />
(i.e., 40 years ago), a paper on the effect of permafrost<br />
and ice wedges on riverbank erosion was presented<br />
(Walker and<br />
Arnborg 1966). Measurements collected included the<br />
depth (horizontal) of thermo-erosional niches (a term borrowed<br />
from the Russian termerosionaja niza) in the permafrost-bound<br />
banks of the delta’s distributaries. Additional<br />
surveys were made of the included ice wedges<br />
along which block collapse occurred (Fig. 1) and benchmarks<br />
were established to monitor thaw rate of bank retreat.<br />
These stations were occupied periodically for the<br />
next 30 years. A number of seasons of measurements<br />
show that the rate of bank retreat varies with bank composition,<br />
bank height, and exposure to erosional processes<br />
(Walker 1983).<br />
Lakes within the delta are highly variable in size,<br />
shape, and origin. All are affected by permafrost to some<br />
extent. Those sufficiently deep have a talik beneath but<br />
most are shallow and floored by permafrost with only a<br />
thin active layer developing during the summer. The most<br />
numerous bodies of water are those present in Iowcentered<br />
polygons bordered by ice wedges (Fig. 2).<br />
Possibly the most unique of the lakes affected by permafrost<br />
in the delta are those perched in the sand dunes.<br />
Perched lakes can only continue to exist so long as there<br />
is an aquaclude between the lake’s water table and the<br />
regional water table; in the case of the Arctic this aquaclude<br />
is often permafrost. Drainage of these ponds occurs<br />
by overflow as the accumulated snow melts and by flow<br />
through the surrounding sand as the active layer develops.<br />
The maximum thickness of the active layer within the<br />
dunes of the Colville delta by the end of the summer is<br />
generally less than two metres. Thus, those ponds in interdune<br />
depressions and in blowouts that are more than two<br />
metres deep retain water throughout the year (Walker<br />
<strong>20</strong>02).<br />
Morphologically, lake tapping provides some of the<br />
most sudden and drastic changes in the delta. Both lake<br />
expansion and river-channel migration lead to tapping<br />
which usually occurs between polygons because the ice<br />
wedges thaw faster than the surrounding permafrost.<br />
Once tapping occurs, lakes fluctuate in level with stage. At<br />
179
low stage lake depths are reduced sufficiently so that Walker, H. J. <strong>20</strong>02. Landform development in an arctic delta; the roles<br />
permafrost begins to form in bottom materials. They also of snow, ice and permafrost. In M. K. Hewitt et al. (eds), Landscapes<br />
of Transition, Dordrecht: Kluwer Publishers: 159-183.<br />
become subject to deposition during the river‘s flood stage<br />
and deltas form at the lake’s entrance.<br />
Sixty years ago, Muller emphasized the engineering<br />
problems associated with permafrost. During most of the<br />
subsequent period of time, most of the permafrost-related<br />
research in the Colville delta was basic. However, in the<br />
199Os, with the exploration for and development of the oil<br />
industry in the delta, research became more applied. It<br />
was aimed at providing ‘I... information essential for engineering<br />
design and evolution of potential environmental<br />
impacts ...” (Jorgenson et al. 1998). Such research, which<br />
is being conducted by a variety of research teams, is providing<br />
much more detailed information about the role of<br />
permafrost in the formation and modification of delta<br />
forms.<br />
REFERENCES<br />
Jorgenson, M. T., Shur, Y. L., and Walker, H. J. 1998. Evolution of a<br />
permafrost-dominated landscape on the Colville River delta,<br />
Northern Alaska. <strong>Proceedings</strong>: Seventh International Permafrost<br />
Conference, Centre d’etudes nordiques, Universite Laval: 523-<br />
529.<br />
Muller, S. W. 1947. Permafrost or Permanently Frozen Ground and<br />
Related Engineering Problems. Ann Arbor: J. W. Edwards Press.<br />
Walker, H. J. and Arnborg, L. 1966. Permafrost and ice-wedge effect<br />
on riverbank erosion. <strong>Proceedings</strong>: First International Permafrost<br />
Conference, National Academy of Science, National <strong>Research</strong><br />
Council, Washington D. C, Publication 1287: 164-171.<br />
Walker, H. J. 1983. Erosion in a permafrost-dominated delta. <strong>Proceedings</strong>:<br />
Fourth International Permafrost Conference. National<br />
Academy of Science Press, Washington D. C.: 1344-1349.<br />
180
Establishing four sites for measuring costal cliff erosion by means of<br />
terrestrial photogrammetry in the Kongsfjorden area, Svalbard<br />
B. Wangensteen, T. Eiken, J. 1. Solid, *R. S. 0degArd<br />
Department of Physical Geography, University of Oslo, Norway<br />
*Gjovik University College, Norway<br />
The Kongsfjorden area is situated at the western coast of<br />
the Svalbard archipelago (Fig. I). The area has continuous<br />
permafrost with measured depths around 140 meters<br />
near the shores of Ny-Alesund (Liestarl 1976). The mean<br />
annual air temperature in Ny-Alesund is -5.8" C (1975 to<br />
<strong>20</strong>00) (KetiI Isaksen, pers. comm.). The tidal range is<br />
about 2 meters. The shores of the area are a mixture of<br />
cliffs in unconsolidated material, cliffs in bedrock and glacier<br />
fronts terminating in the sea. The selected four sites<br />
consist of low cliffs in bedrock and unconsolidated material,<br />
both with stony beaches in front (0deg5rd et al.<br />
1987). The geology at three of the sites is lime- and<br />
dolostone (Carbon and Perm) while one site is in an area<br />
of gneiss (upper Proterozoic). The unconsolidated beach<br />
material is of Quaternary origin (Hjelle 1993). The area<br />
has earlier been subjected to some related research activities<br />
concerning coastal processes e.g., cryogenic<br />
weathering of cliffs (0degard and Sollid 1993).<br />
At all sites a fixed point for global reference was established;<br />
a bolt was drilled into the bedrock and its position<br />
measured by GPS. The positions of the camera stations<br />
were measured by GPS and surveyed from the fixed point<br />
to create both a global and a local reference. The photos<br />
were acquired with a stereo overlap using a Hasselblad<br />
camera with a 60 mm lens. At two of the sites bolts to be<br />
used as photogrammetric control points were drilled into<br />
the cliff wall as well. At the sites consisting of<br />
unconsolidated material no control points were established.<br />
The distance between the camera stations and the<br />
cliff wall varied between 7 and 15 meters for the four sites.<br />
The distance between the camera stations was approximately<br />
half of the photo distance. The coordinate system<br />
is transformed in a manner to simulate the geometry of<br />
aerial photogrammetry. The z-axis is pointing out of the<br />
cliff and the xy-plane is parallel to the cliff wall and also<br />
the line between the photo stations. The camera was<br />
mounted on top of the theodolite with a special device<br />
making it possible to measure the exact position of the<br />
camera and to ensure that the photo stations are on line<br />
parallel to the cliff wall. The camera positions are also<br />
used in the photogrammetric orientation of the scenes. For<br />
the scenes from the sites of unconsoli dated material this<br />
set-up also gives the opportunity to measure the photo<br />
Figure 1. Map showing location of the investigated area.<br />
angle used for exterior orientation of the scenes. This is<br />
important since no control points are available. The photos<br />
were scanned at a resolution of 11 pm (2,300 dpi) and<br />
imported into the Zll Imaging digital photogrammetric<br />
workstation software. To be able to use the Zll Imaging<br />
software the coordinate system had to be rotated, as explained<br />
earlier, and all coordinates up-scaled by a magnitude<br />
of 100, both to simulated aerial photography. After<br />
the photo pair or triples have been used to create a photogrammetric<br />
stereo model it is possible to automatically<br />
generate a high-precision digital terrain model (DTM). The<br />
Match-T algorithm in Zll Imaging utilizes cross-correlation<br />
to detect the same points in the different photos of the stereo<br />
model and then measures the elevation by measuring<br />
the y-parallax. Photo distances in the range of 7 to 15 m<br />
give a scale of 1 :I 17 to 1 :<strong>25</strong>0. With the above-mentioned<br />
scanning resolution the XY-resolution is in the order of 1.2<br />
to 2.5 mm. The accuracy of elevation measurements done<br />
by photogrammetry is about 0.<strong>20</strong> to 0.40%0 of the photo<br />
distance. This gives a theoretical elevation accuracy (Zdirection)<br />
of 1.4 to 2.8 mm for the 7 m photo distance and<br />
3.0 to 6.0 mm for the 15 m photo distance.<br />
-4<br />
181
The automatically-generated DTM for one of the sites<br />
is shown in Figure 2 as contour lines on top of an orthophoto<br />
of the cliff wall. The height and width of the area<br />
covered by the model is 3.3 m and 2.8 m respectively.<br />
When investigated in the digital photogrammetric workstation,<br />
the terrain model follows the terrain quite nicely. Precision<br />
estimates within the Zll Imaging software gives a<br />
value of 1 to 2 mm for the Z-accuracy at this site. The<br />
photo distance used here was 7 meters.<br />
REFERENCES<br />
And& M.-F. 1997. Holocene Rockwall Retreat in Svalbard: A triplerate<br />
evolution. Earth Surface and Processes and Landforms 22:<br />
423-440.<br />
Hjelle, A. 1993. Geology of Svalbard. Polarhhdbok No. 7, Norsk Polarinstitutt,<br />
Oslo.<br />
Liestol, 0. 1976. Pingos, springs and permafrost in Spitsbergen. In<br />
Norsk Polarinstitutts Arbok 19757-29.<br />
0degArd, R., Sollid, J.L. and Trollvik, J.A. 1987. Coastal Map Svalbard<br />
A 3 Forlandssundet 1:<strong>20</strong>0.000. Department of Physical Geography,<br />
University of Oslo.<br />
0degArd, R. Sollid, J.L. 1993. Coastal cliff temperatures related to the<br />
potential for cryogenic weathering processes, western Spitsbergen,<br />
Svalbard. Polar <strong>Research</strong> 12(1): 95-106.<br />
Fj~ur~ 2. ~ont~ur lines of 5 crn<br />
~ha~a. The size of the rno~~i ar<br />
~ i ~ ~ ~ ~ h ~ ~ ~ ~ ~ i ~ ~ l ~ ~ .<br />
The method of close up digital terrestrial photogrammetry<br />
seems suitable for creating accurate digital terrain<br />
models of coastal cliffs and hence promising for calculating<br />
erosion rates in the order of mmlyear. The intention is<br />
to acquire a new set of photos of the same sites in <strong>20</strong>04 to<br />
generate new terrain models that will enable the erosion<br />
rate calculations. Andre (1997) reports a rock wall retreat<br />
in the range 0 to 1.58 mmlyear in the same area. The<br />
coastal cliff erosion rate is unknown but believed to be<br />
higher than the rock wall retreat rate and hopefully greater<br />
than the accuracy of the technique used.<br />
The fieldwork was financially supported by the INTASproject<br />
“Arctic coasts of Eurasia: dynamics, sediment<br />
budget and carbon flux in connection with permafrost<br />
degradation’’ (INTAS-<strong>20</strong>01-2329) and the Norwegian<br />
<strong>Research</strong> Council on behalf of the Norwegian Polar<br />
Committee.<br />
182
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Av~KI~~Ie Information<br />
Laboratory experiments towards better understanding of surface buckling<br />
of rock glaciers<br />
M. Weber and A. Kaab<br />
Glaciology and Geomorphodynamics Group, Department of Geography, University of <strong>Zurich</strong>, <strong>Switzerland</strong><br />
Wahrhaftig and Cox (1959) assumed permafrost creep<br />
to be the main transfer process in rock glacier flow. A<br />
number of boreholes on active rock glaciers give insight<br />
into material properties, rheology and the thermal conditions<br />
involved (Haeberli et al. 1998). Information on<br />
kinematics of rock glaciers has been obtained from terrestrial<br />
measurements and computer-aided aerial photogrammetry<br />
(Kaab 1998). Numerical modelling showed<br />
that transverse ridges might develop after a change in<br />
slope if two layers with different viscosity exist (Loewenherz<br />
et al. 1989).Yet still many questions concerning the<br />
interaction between kinematics and structure of a rock<br />
glacier remain unanswered. The characteristic topographic<br />
surface structures on rock glaciers, namely transverse<br />
ridges, can be regarded as a physical expression<br />
of these interactions. Within the framework of a diploma<br />
thesis aiming to analyze the occurrence of surface<br />
structures in combination with surface parameters, a<br />
creeping mass has been physically modelled in a laboratory<br />
experiment to qualitatively estimate possible<br />
processes involved in such surface buckling. The study<br />
addresses the following questions: is there a suitable<br />
substitute for the debris-ice mixture of a real rock glacier<br />
to be used in laboratory simulations Is a decrease in<br />
longitudinal slope sufficient to trigger the evolution of<br />
transverse ridges in a viscous material If not, which<br />
material properties allow transverse ridges to develop<br />
In our experiments the food additive Xanthan Gum<br />
(E415) has been tested. Xanthan gum is a microbial<br />
polymer prepared commercially in a pure culture fermentation<br />
of the bacterium Xanthomonas campestris.<br />
The molecular structure of Xanthan Gum results in<br />
pseudo plastic flow behaviour. Xanthan Gum is used in<br />
food engineering as well as cosmetics, and the petrochemical<br />
industry in order to control flow processes. Our<br />
experiments were carried out on a ramp built from perspex.<br />
The ramp is constructed such that its slope can be<br />
varied. (The discontinuity between the ramp and the<br />
horizontal base is referred to here as the "break of<br />
slope"). A container is fixed to the top of the ramp, used<br />
to provide equal starting conditions for each experiment.<br />
In changing the mixture of two compounds differing in<br />
viscosity and in solids content such as sand andlor<br />
gravel, the rheological behaviour of the material has<br />
been varied in a systematic way.<br />
In summary, the experiments showed the following<br />
trends:<br />
- No or only very minor transverse ridges at or near the<br />
break of slope are created if the material was well<br />
mixed and "homogeneous" (Le., no solids content).<br />
- If two components of different viscosities but no solids<br />
content are only slightly mixed ("heterogeneous"),<br />
fine transverse ridges could develop near the break<br />
of slope. After sufficient material has crept downwards<br />
and starts to accumulate in the horizontal,<br />
ridges develop above the break of slope.<br />
- Ridges also develop if sand andlor gravel are added<br />
to a single-component mass.<br />
- If a mass comprised of sand- and gravel-containing<br />
parts of different viscosity is only slightly mixed or<br />
applied in two separate layers, distinctive ridges appear.<br />
- In another experiment, where a slightly mixed mass<br />
with sand and gravel ("heterogeneous") formed the<br />
bottom layer covered by a dry gravel brittle top layer,<br />
distinctive ridges developed. Moreover, due to the<br />
compression at the point of break in slope, single<br />
gravel particles were dislocated and fell into the<br />
newly-formed furrows. Viscous material was pushed<br />
from underneath to the surface forming ridges of increasing<br />
amplitude.<br />
The findings from these experiments exhibit analogies<br />
to real rock glaciers. Although the observed processes<br />
are certainly not equivalent to those occurring in<br />
nature, the experiments yield structures comparable to<br />
those of real rock glaciers. A development model for<br />
transverse ridges can be deduced from the experiment<br />
described above in which a brittle layer has been forced<br />
up through viscous material. According to this model, it<br />
may be possible that transverse ridges on rock glaciers<br />
can develop due to the compressing flow of the viscous<br />
ice-debris-mass and its interaction with an overlying brittle<br />
layer (i.e., the active layer). It may also be possible
due to the entire flow process. In any case, a heterogenous<br />
variation of viscosities, or solids content, respectively,<br />
seems to favour or even be a prerequisite for<br />
transverse surface structures to develop. Such conceptual<br />
models are presently being investigated in relevant<br />
high-precision field surveys.<br />
ACKNOWLEDGEMENTS<br />
Special thanks to STAERKLE & NAGLER AG, Zollikon,<br />
<strong>Switzerland</strong> for providing Xanthan Gum.<br />
Weber, M. et al.: Surface buckling of rack glaciers,..<br />
REFERENCES<br />
Haeberli, W. 1985. Creep of mountain permafrost: internal structure<br />
and flow of alpine rock glaciers. Mitt. Versuchsanst. Wasserbau<br />
Hydrologie Glaziologie ETH <strong>Zurich</strong>. 77 pp.<br />
Kaab, A., Gudmundsson, G.H. and Hoelzle, M. 1998: Surface deformation<br />
of creeping mountain permafrost. Photogrammetric<br />
investigations on Murtel Rock Glacier, Swiss Alps. <strong>Proceedings</strong><br />
of the Seventh International Conference on Permafrost, Yellowknife,<br />
Canada, Collection Nordicana 57: 531-537.<br />
Loewenhen, D. S., Lawrence, C. J., Weaver, R. L. 1989: On the<br />
development of transverse ridges on rock glaciers. Journal of<br />
Glaciology 35 (121): 383-391,<br />
Wahrhaftig, C., Cox, A. 1959: Rock Glaciers in the Alaska Range.<br />
Bulletin of the Geological Society of America 70: 383-436.<br />
Weber, M. <strong>20</strong>03: Oberflachenstrukturen auf Blockgletschern. Diploma<br />
thesis, University of <strong>Zurich</strong>.<br />
184
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Russian Information Transfer Programme<br />
P. J. Williams and 1. M. T. Warren<br />
Scott Polar <strong>Research</strong> Institute, Cambridge, United Kingdom<br />
The longest tradition of permafrost knowledge and the<br />
largest body of accumulated information is Russian. Russian<br />
studies can be traced back centuries, while it was<br />
only in the second world war that permafrost became of<br />
broad practical interest and an object of much scientific investigation<br />
in North America. In Finland, Norway and Sweden<br />
with their cold climates, and elsewhere, occasional<br />
studies have been made over two centuries at least. But<br />
apart from the contributions of several key figures, for example<br />
Taber, Beskow or Troll, only in Russia were there<br />
large-scale organised studies of frozen ground and permafrost<br />
and the corresponding technologies, through the<br />
first half of the twentieth century.<br />
This Russian contribution continues and will increase,<br />
considering how many people live in a permafrost environment<br />
there, the extent of industrial (oil, gas, mineral<br />
development) and other human activities (including those<br />
of many native peoples). Development of Alaska and its<br />
natural resources and of northern Canada has brought<br />
similar intense interest in the scientific, industrial, geotechnical<br />
and general environmental conditions of the<br />
permafrost regions there. Such is modern society, permafrost<br />
is of global significance.<br />
Most permafrost literature is in Russian or English, both<br />
major languages but not, for most individuals, interchangeable.<br />
The difficulties concern engineers, financiers,<br />
politicians and many others now involved with the cold regions.<br />
Accessibility to the information, however, is of the<br />
utmost importance - it may be extremely costly in financial<br />
and in human terms, to remain unaware of the hard-won<br />
expertise of others.<br />
The Russian Information Transfer Programme of the<br />
Scott Polar <strong>Research</strong> Institute in Cambridge, England, is<br />
designed to bring the significance of ‘permafrost‘ literature<br />
in the Russian language to those who cannot read it. It<br />
goes further than translation: it is a scholarly undertaking<br />
to ensure that the Russian work is as meaningful to English-<br />
language readers as to the Russian authors themselves.<br />
The history of the twentieth century was such that<br />
studies relating to Siberia, to the oil and gas industry, to<br />
185<br />
military interests, and to the cold regions generally were<br />
not widely revealed to those outside the Soviet Union.<br />
Even more important was that the general lack of scientific<br />
exchange resulted in the development of concepts and<br />
models distinctly Russian on the one hand and distinctly<br />
‘English language’ or ‘Western’ on the other. And however<br />
etymologically-polished a translation may appear, it loses<br />
much of its value if it is not quite understood by the specialist<br />
reader. Think merely of ‘cryolithological regions’,<br />
‘massive permafrost’, ‘basal ice’- terms quite acceptable to<br />
a first-rate translator but full of uncertainty to the permafrost-sawy<br />
English language reader.<br />
The first undertaking under the nascent Programme<br />
was Professor Yershov’s Obshchaya Geokriologiya -<br />
General Geocryology (1998), which is a comprehensive<br />
overview of Russian permafrost science and technology.<br />
The aim was to publish the book to a comparable standard<br />
to that if had been authored in English. This was a<br />
challenge but, even if we were not completely successful<br />
in meeting it, the book has given to English language<br />
readers a unique insight into the depth of Russian knowledge<br />
and its implications.<br />
Building on this experience, an English-language folio<br />
volume was prepared to accompany the original maps of<br />
the 16-sheet highly informative Geocryological Map of<br />
Russia and Neighbouring Republics. All legends are<br />
translated, with additional explanations and a glossary.<br />
The Map set has immediate practical applications in planning,<br />
and for major engineering and other works. It gives<br />
insight into permafrost distribution and origin, active layer<br />
conditions and many other features, as well as into the<br />
manner in which such information is interpreted and utilised.<br />
The second edition (Williams and Warren <strong>20</strong>03) of<br />
this folio volume (together with the maps) is being published<br />
shortly to meet the increased global interest in oil<br />
and gas especially, and in other aspects of the Russian<br />
cold regions. It includes refinements of the glossary, the<br />
translations and other components of the volume. Both<br />
Yershov’s book and the Map have led to extensive discussions<br />
of interpretation with numerous specialists.
Williams, P. J. et al.: Russian Information Transfer Prograrnrne<br />
The next major undertaking has been the preparation<br />
of a volume of selected articles from Kriosfera Zemli -<br />
Earth’s Cryosphere, the leading Russian journal devoted<br />
to permafrost and related topics. This work too has been<br />
carried out as a scholarly undertaking and, because of the<br />
diversity and importance of topics covered (for example,<br />
electrical and mechanical properties of frozen ground;<br />
permafrost distribution in coming years; gas hydrate stability;<br />
geotechnical foundation techniques; biological activity<br />
in permafrost) the advice of even more specialists<br />
(some never before involved with permafrost), has been<br />
sought. The aim is to continue to publish an English language<br />
version of this unique journal - but this will depend<br />
on the degree of financial support achieved through sales<br />
and other sources.<br />
The Programme has benefited from a generous, collaborative<br />
approach from both the Russian and Englishlanguage<br />
sides, and especially from the authors, editors<br />
and publishers of the Russian studies. The Programme<br />
has led to new linkages and has given those working with<br />
it a fuller bibliographic insight, as well as a depth of understanding<br />
of the people and organisations that are contributing<br />
to our knowledge of geocryology. Consequently the<br />
Programme is also able to make other contributions in the<br />
form of advice, and, when funds are available, of literature<br />
searches and analyses and ‘custom’ translations. European<br />
Union funds have been awarded for visits to meetings<br />
by Russian specialists, especially relating to pollution<br />
in permafrost regions, strengthening the Programme in<br />
this topic.<br />
Political developments and the significance of the permafrost<br />
regions point to a growth in coming years of the<br />
Russian Information Transfer Programme, but this will depend<br />
on the response of those who utilise it. Most agencies<br />
funding research doubtless see these activities as<br />
being peripheral to their responsibilities. Yet it is cheaper<br />
to benefit by the experience of others than to rediscover<br />
the problems and the expense of uncertain solutions, especially<br />
when the costs of making such knowledge available<br />
are small indeed compared to the financial and other<br />
dangers of proceeding in ignorance.<br />
i<br />
REFERENCES<br />
Yershov, E. 1998. General Geocryology: (originally published as<br />
Obshchaya Geokriologiya, Nedra, Moscow). Cambridge University<br />
Press, 580pp ISBN 0 521 47334 9.<br />
Williams, P.J. and Warren I.M.T. <strong>20</strong>03.The English Version of the<br />
Geocryological Map of Russia and Neighbouring Republics.2nd.<br />
Edn. Original maps, guide, folio, I-X, & 21pp. Cambridge University,<br />
Carleton University, University of Moscow. Ottawa: Collaborative<br />
Map Project. ISBN 0-9685013-0-3.<br />
186
8th international Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Permafrost as a natural cryobank of late Pleistocene and modern plant<br />
germplasm<br />
Institute of Cell Biophysics, Russian Academy<br />
S. G. Yashina, E .N .Gakhova and *S. V. Gubin<br />
of Sciences, Pushchino, Moscow Region, Russia<br />
"Institute of Physicochemical and Biological Problems in Soil Science, Russian Academy of Sciences, Pushchino, Moscow<br />
Region, Russia<br />
INTRODUCTION<br />
RESULTS AND<br />
DISCUSSION<br />
Conserving the world's biological diversity has recently become<br />
a global problem. The scientific community is now<br />
ready to use the advances in cryobiology to initiate long-term<br />
strategic and sustainable genetic resource cryobanks for<br />
saving endangered species, in addition to the traditional<br />
means such as nature reservation, botanical gardens, etc.<br />
(Gakhova 1998, Tikhonova and Yashina 1998). In recent<br />
years, the viable spores of mosses (Gilichinsky et al. <strong>20</strong>01)<br />
and seed embryo cells of the higher plants (Yashina et al.<br />
<strong>20</strong>02) have been discovered within Siberian late Pleistocene<br />
permafrost. It was demonstrated that they were able to de<br />
velop and grow. In connection with this, new prospects have<br />
been revealed for cryopreservation and long-term banking<br />
germplasm of the modern wild plant. In this study we show<br />
current research on vital functions of higher plant seeds from<br />
permafrost deposits and present some data for the use d<br />
permafrost for the conservation of modern plant seeds.<br />
MATERIALS AND METHODS<br />
The seeds and unripe fruits were evoked from the fossil barrows<br />
of ground squirrel in the late Pleistocene ice-loess<br />
sediments of North Yakutiya (Gubin and Khasanov 1996).<br />
The burrows were found in the constant annual subzero temperature<br />
layer at a depth of IOto 18 m. Radiocarbon age d<br />
the paleontological material was determined to be an average<br />
of 30 thousand years. Seeds were identified by means of a<br />
collection of plant seeds from the Kolymsko-lndigiskaya<br />
lowland (Yashina et al. <strong>20</strong>02). The fossil seeds were into<br />
duced into in vitro culture on an agar nutritious medium to ascertain<br />
their vital capacity. Methods of plant tissue culture and<br />
clonal micropropagation of plant in vitro, light microscopy and<br />
microphoto were used.<br />
The long-term experiments on storage of modern wild<br />
plant (14 species) seeds were performed in a permafrost<br />
zone at various temperature conditions. Viability of modem<br />
plant seeds was determined by periodic control over the lab<br />
ratory germinating capacity.<br />
Seeds of sufficient diversity were found in the burrow studied<br />
(up to IOspecies). Attempts at seed germination were undertaken<br />
for each species, but physiological viability of seeds<br />
was discovered only in 3 species: stenophyllous campion<br />
(Silene stenophylla Ledeb.) of the pink family (Caryophyllaceae)<br />
and 2 species, identified accurate to the genus: arctous<br />
(Arcfous spp.) of the heather family (Ericaceae) and<br />
persicaria (Polygonum spp.) of the persicaria family (Polygonaceae).<br />
The plantlet of persicaria seed has germinated in vifro<br />
culture to cotyledon leaf stage (Fig. I). Callusing from eE<br />
dospem cells of arctous seeds has been obtained also in<br />
culture. It tumed out that campion seeds are more tolerant b<br />
long-term storage at a subzero temperature. The initial stage d<br />
embryonal root growth and callus formation from the resulting<br />
roots cells has been induced on several of campion seeds.<br />
Development of embryonal roots has not occurred because d<br />
degradation of the storage tissue. At the same time, fragments<br />
of embryonal tissue in unripe fruits of campion have been<br />
shown to have the properties of enlargement (Fig. 2) and organogenesis<br />
formation. At present, the plant material obtained<br />
provides the basis for clonal micropropagation and the subseguent<br />
recovery of plants of full value. The long-standing experiments<br />
on storage of wild plant seeds are performed in a<br />
permafrost zone at different temperature regimes in order b<br />
use permafrost for conservation of the plant genetic resources.<br />
Part of the seed samples were placed into the glacier<br />
(Kolymsko-lndigiskaya lowland) at mean annual temperatures<br />
of -15OC, the rest were placed into an underground<br />
depository in the Melnikov permafrost Institute of SD RAS<br />
(Yakutsk) and stored at -2,7OC. <strong>Research</strong> on viability of<br />
seeds by determination of laboratory germination capacity<br />
over 1.5, 2 and 5 years have shown that glacier conditions<br />
had some advantages over the underground storage d<br />
seeds. At thesametimetheseedsofsomespecies had<br />
more high germination capacity in underground storage in relation<br />
to control (laboratory storage at +4OC). It is interesting<br />
that microbiotic seeds (Anemone sylvestris L., Pulsatilla patens<br />
(L.) Mill.) preserved high viability in the glacier as well<br />
as in the underground depository, whereas these seeds rag<br />
idly lost germination capacity at +4OC. Both easily germinat-<br />
187
ing seeds (Dianthus fischeri Spreng, D. superbus L., Leucanthemum<br />
vulgare Lam.) and seeds with biological dormancy<br />
(Liliurn martagon L., Lavafera thuringiaca L.) had high<br />
viability after storage in the glacier. Seeds of several species<br />
in the pink family demonstrated especially high viability after<br />
storage in the glacier for 5 years. It is noteworthy that the viable<br />
fossil stenophyllous campion belongs to this same family.<br />
Yashina, S. G. et al.: Permafrost as a natural cryobank ...<br />
ACKNOWLEDMENT<br />
This work was supported by the Russian Foundation of Basic<br />
<strong>Research</strong> (grants 99-0449369, 01-0449087 and 0244-<br />
49369).<br />
REFERENCES<br />
Fi~ure 1. The ~lantlet of fossil ~ ~~si~aria seed<br />
Gakhova, E.N. 1998. Genetic cryobanks for conservation of biodiversity.<br />
The development and current status of this problem in<br />
Russia. Cryo-Letters, suppl.1: 57-64.<br />
Gilichinsky, D.A. et al <strong>20</strong>01. How long the life might be preserved<br />
In (Giovannelli F., ed.), <strong>Proceedings</strong> of the International Workshop<br />
((The Bridge between the Big Bang and Biology)), Consiglio<br />
Nazionale delle Ricerche of Italy: 362-369.<br />
Gubin, S.V. and Khasanov, B.R. 1996. The fossil barrows of Mammalia<br />
from permafrost sediments of Kolymsko-lndigirskaya<br />
Lowland. Dokl.Akad.Nauk 346(2): 278-279 (in Russian).<br />
Tikhonova, V.L. and Yashina, S.G. Long-term storage of endangered<br />
wild plant seeds. Physiol. Gen. Biol. Rew. 13(1): 1-33.<br />
Yashina, S.G. et al. <strong>20</strong>02. Viability of higher plant seeds of late<br />
Pleistocene age found in permafrost deposits as determined by<br />
in vitro culturing. Dokl.Akad.Nauk 383(5): 714-717 (in Russian).<br />
Figure 2. The fossil campion seed with fragment<br />
bryonal tissue<br />
of enlarging em-<br />
CONCLUSION<br />
For the first time, it has been demonstrated that cells of fragments<br />
of unripe fruits of stenophyllous campion and seed embryo<br />
cells of this species, as well as seed embryo cells d<br />
two unidentified species of the genera arctous and polygonum,<br />
preserved viability during thirty thousand years in<br />
permafrost. Physiological activity of seed cells has been discovered<br />
by the in vitro cultural method. This pioneering<br />
method for the conservation of seeds of modern wild species<br />
under permafrost conditions may be promising because this<br />
type of permafrost storage is protected from natural disasters<br />
and anthropogenic effects, is relatively inexpensive, and can<br />
maintain a stable subzero temperature. Many importance<br />
questions remain to be resolved including the identification d<br />
natural mechanisms of adaptation that allow plant cells and<br />
tissues to preserve vital capacity for thousands of years at<br />
subzero temperatures. Also, there is a need to study tern<br />
perature regimes and other physicochemical conditions the for<br />
long-term storage of plant germplasm in permafrost.<br />
188
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Active layer monitoring in Northeast Russia: regional and interannual<br />
variability<br />
D. Zamolodchikov, *A. Kofov, **D. Karelin and ***V. Rarzhivin<br />
Centre for Ecology and Productivity of Forests, Russian Academy of Sciences, Moscow, Russia<br />
The CALM network (Brown et al. <strong>20</strong>00) presently includes<br />
three sites in northeast Russia (Chukotka district). The<br />
first site, Cape Rogozhny, was established in I994 on the<br />
northern coast of Onemen Bay (64"49'N, 176'50'E) of the<br />
Bering Sea. The second site, Mount Dionisy, started in<br />
I996 and is located about <strong>25</strong> km south of Anadyr city<br />
(64"34'N, 177'12'E). Both sites are dominated by cottongrass<br />
tussocks on Gleyi-Histic Cryosols. Frost cracks and<br />
frost boils are well developed. The third site, Lavrentia,<br />
was initiated in <strong>20</strong>00 in the vicinity of Lavrentia settlement<br />
(65036'N, 171°03'W). The site consists of wet sedge-Salix<br />
mosses tundra without developed microrelief on Gleyi-<br />
Histic Cryosols. This report was developed as a result of<br />
the CALM synthesis workshop held in November <strong>20</strong>02 at<br />
the University of Delaware.<br />
Grids (100x100 m) are used for active layer monitoring<br />
at the above sites. Thaw depth is measured using a steel<br />
rod (121 points per grid). The surface soil moisture is<br />
monitored at the Lavrentia site with a Vitel hydra probe<br />
(Hydra soil moisture probe 1994). Only end-of-season<br />
data are available for Cape Rogozhny and Mount Dionisy.<br />
Seasonal measurements were done at the Lavrentia site<br />
(3-4 point per summer).<br />
The climate conditions at Cape Rogozhny and Mount<br />
Dionisy are well described by data from the Anadyr<br />
weather station (64"47'N, 177'34'E). The closest weather<br />
station to the Lavrentia site is at the Uelen settlement<br />
(6601 O'N , 169050'W).<br />
The mean annual air temperature at the Anadyr and<br />
Uelen stations weather stations are fairly similar (Fig. 1A)<br />
in absolute values and interannual dynamics (R=0.89).<br />
Some temperature increase is observed from I999 to<br />
<strong>20</strong>02, but the warmest period was 1995 to 1997. For the<br />
period of CALM observations (1994-<strong>20</strong>02), a trend in the<br />
mean annual temperature is not noticeable.<br />
An important climatic predictor of seasonal thawing is<br />
thawing degree days (DDT) value (Brown et al. <strong>20</strong>00). The<br />
DDT values are different in the Anadyr and Uelen stations<br />
(Fig. IB). The difference is explained by colder summers<br />
in the more marine conditions of Uelen and Lavrentia locations.<br />
The warmer winters at the above locations lead to<br />
similarities in the annual temperature regimes.<br />
*Scientific Center "Chukofka': Anadyr, Russia<br />
**Biological department, Moscow State University, Moscow, Russia<br />
***Komarov Botanical Institute, St-Petersburg, Russia<br />
1994 1995 1996 1997 1998 1999 <strong>20</strong>00 <strong>20</strong>01 <strong>20</strong>02<br />
<strong>20</strong>0 I<br />
1994 1995 1996 1997 1998 <strong>20</strong>00 1999 <strong>20</strong>01 <strong>20</strong>02<br />
3s 1<br />
_I" 1<br />
- - - L, - - Cane Ronozhnv<br />
"a- Mount of Dionisv<br />
1994 1995 1996 1997 1998 1999 <strong>20</strong>00 <strong>20</strong>01 <strong>20</strong>02<br />
Year<br />
Figure 1. Dynamics of (A) mean annual air temperature, (B) DDT, and<br />
(C) end-of-season thaw depth at the CALM sites in northeast Russia.<br />
The mean end-of-season thaw depth in Cape Rogozhny<br />
is 44 cm with interannual variations from 38 to 50 cm<br />
(Fig. IC). The peak of seasonal thawing was registered in<br />
1996 and 1997 (48 and 50 cm respectively).<br />
The mean end-of-season thaw depth at Mount Dionisy<br />
is 49 cm with interannual variations of 45 to 53 cm. The<br />
most northern CALM site (Lavrentia) is characterized by<br />
the largest end-of-season thaw depth (60 cm), with small<br />
interannual variations (59 to 61 cm).<br />
End-of-season thaw depth at Cape Rogozhny is correlated<br />
(R=0.62) with the square root of DDT (Fig. 2). The<br />
correlation is more pronounced (R=0.71) for thaw depth<br />
189<br />
b-*k
and mean annual temperature. This result stresses the<br />
role that warm winters play in determining frozen ground<br />
temperatures and corresponding lower energy requirements<br />
for the thawing season.<br />
The correlation of end-of-season thaw depth and<br />
DDTo.5 is poor at the Mount Dionisy site (Fig. 2). This has<br />
two reasonable explanations. The first is related to the<br />
early dates of measurements (near 10 August) in some<br />
years (1 997, <strong>20</strong>00). Actual end-of-season thaw depth can<br />
be different from the measured values in these years.<br />
The second is due to landscape position of the site where<br />
the underground water flows determine the spatial patterns<br />
of soil thawing. Under these conditions the actual<br />
thaw depth is determined more by summer precipitation<br />
than by air temperatures.<br />
The end-of-season thaw depth for the Lavrentia site is<br />
well-correlated with DDT0.5 (R=0.96). However, the time<br />
series are limited for the site (n=3).<br />
Analysis of the current CALM grid data allows us to<br />
characterize the main influences of spatial variability on<br />
thaw depth. At the Cape Rogozhny site the above factors<br />
are disturbances of vegetation and soil cover (frozen boils,<br />
tracks of all-terrain vehicles). In the Mountain Dionisy site,<br />
the spatial variability of the active layer is determined by<br />
subsurface water flows, corresponding to depressions of<br />
mesorelief. Frost boils also input to active layer variability<br />
at this site. The main control factor of end-of-season thaw<br />
depth at the Lavrentia site is the thickness of the soil organic<br />
layer.<br />
Seasonal pattern of measurements at the Lavrentia site<br />
make it possible to reveal the changes of spatial variability<br />
control factor during soil thawing. At the beginning of the<br />
warm season, the thickness of organic soil layer, surface<br />
soil moisture and thickness of moss cover demonstrate<br />
approximately the same relation to thaw depth (absolute<br />
values of R coefficients vary from 0.3 to 0.4). In the course<br />
of the season the role of organic layer increases (R<br />
achieves -0.6), while the input of other factors decreases<br />
(absolute value of R diminishes to 0.2).<br />
The present review of regional CALM studies in northeast<br />
Russia results in the following conclusions:<br />
there are no evident trends in end-of-season thaw<br />
depth in northeast Russia;<br />
drainage ways, surface disturbances and the thickness<br />
of the organic soil horizon have a major influence<br />
on the spatial variations of the end-of-season<br />
thaw depth;<br />
the influence of different factors on the thawing<br />
process is seasonally specific.<br />
TDL = 1.412DDT0'5 + <strong>20</strong>. I 1<br />
R2 = 0.913 /*<br />
/ TDD = 0.1'3D~o's + 4297<br />
/ 0 R2= 0.043<br />
6<br />
.O"""<br />
_ _ " "<br />
"..,~,"""'"~~""<br />
I"<br />
/ ,<br />
n<br />
'I I<br />
c<br />
R' = 0.392<br />
""<br />
I<br />
-.-I<br />
22 27 32<br />
DD<br />
35 ..<br />
0 CapeRogcahny<br />
0 Mount Dmisy<br />
Lavrmtla<br />
""<br />
""-7 j<br />
. -.<br />
Figure 2. End-of-season thaw depth versus square root of thawing<br />
degree days for three CALM sites in northeast Russia.<br />
REFERENCES<br />
Brown, J., Hinkel K.M. & Nelson F.E. <strong>20</strong>00. The Circumpolar Active<br />
Layer Monitoring (CALM) program: research designs and initial<br />
results. Polar Geography 24(3): 166-<strong>25</strong>8.<br />
Hydra soil moisture probe user's manual. Version 1.2. 1994. Vitel,<br />
Inc., Chantilly, VA, 24 pp.<br />
Zamolodchikov D.G., Karelin D.V., lvaschenko A.I., Oechel W.C &<br />
Hastings S.J. in press. COz flux measurements in Russian Far<br />
East tundra using eddy covariance and closed chamber techniques.<br />
Tellus. Special Issue.<br />
ACKNOWLEDMENT<br />
The fieldwork at the Lavrentia site was supported by the<br />
National Science Foundation (USA), Arctic Systems Science,<br />
Land-Atmosphere-lce-Interactions Program (OPP-<br />
9216109), as part of flux studies in Arctic ecosystems<br />
(Zamolodchikov et al. <strong>20</strong>03).
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Pedogenesis of carga thermochrone paleosols in North-East Russia<br />
0. G. Zanina and S. Vi Gubin<br />
Soil Cryology Laboratory, Insfifute of Physicochemical and Biological Problems in Soil Science, Russian Academy of Sciences,<br />
Pushchino, Moscow Region, Russia<br />
INTRODUCTION<br />
Epigenetic pedogenesis<br />
Paleosols of early Carga pedocomplex. Early Carga pa-<br />
Late Pleistocene soils buried in ice-loess deposits were used leosols were formed during an initial thermochrone stage. The<br />
for reconstructing environments of the Carga thermochrone. age of soils exceeds 37 thousand years. Paleosols of this<br />
The Carga thermochrone ( 50 to 28 thousand years BP) is a period do not have the attributes of wood-soil formation. Their<br />
period of latePleistocenetime. In theNortheastofRussia profiles are well differentiated into genetic horizons. Large<br />
paleosols developed during this period. These paleosols are wood remains are found within humic horizons of these pacharacterised<br />
by a complex organisation; the deposits contain leosols.<br />
epigenetic and synlithogenetic soils (cryopedolite). Epigenetic Paleosols of late Carga pedocomplex. Soil formation took<br />
pedogenesis in the period of the Late Pleistocene was repeat- place during later periods of Carga time, 35 to 28 thousand<br />
edly replaced by synlithogenetic pedogenesis. The terms years ago. Pedocomplexes include three buried soils of dif-<br />
“synlithogenetic” and “epigenetic” are used for the description ferent ages. Structures of paleosol profiles which were formed<br />
of soil developments during this period. Epigenetic soil forma- during the initial stage of pedogenesis within the pedocomplex<br />
tion refers to processes where differentiation of the active<br />
(the first and the second soils) are best developed. Paleosols<br />
layer material takes place in genetic horizons and soil profiles are represented by Histosols and Gleysols, which are didevelop.<br />
Synlithogenetic soil formation occurs when material vided by layers of cryopedolite.<br />
of deposits is transformed by pedogenetic processes in the The organic horizons of Gleysols differ with respect b<br />
absence of soil profiles. The different approaches for studying structure, properties and thickness. In these soils the hydrothese<br />
layers are determined by their genesis.<br />
morphic characteristics and gleying are well expressed. The<br />
thickness of profiles varies between to 1 2 m. The structure d<br />
organic horizons in moss and sedgy peat is different. The up<br />
METHODS<br />
per (third) soil has badly marked structures. It was formed<br />
under drier and even arid conditions of a harsh climate.<br />
Buried soils contain a set of fossil organic materials in their<br />
organic horizons (peatland, humic-peatland). Conditions d<br />
soil formation are reconstructed using traditional paleosol<br />
methods on the basis of analyzed organic remains and of<br />
other soil properties. For studying synlithogenetic soils, detritus,<br />
spores and pollen existing in the deposits are analysed.<br />
Fossil rodent burrows of Carga cryopedolite are most infor-<br />
mative for reconstruction due to their high content in biological<br />
matter.<br />
OBJECTS<br />
Paleosols were investigated in eight outcrops of the lower<br />
course of the Kolyma River. As a result of long-term research<br />
in the Kolyma lowland, buried soils including the two<br />
pedocomplexes - early Carga complex and late Carga<br />
complex - were determined (Gubin 1994). Layers of loess<br />
with attributes of synlithogenetic soil formation are covered by<br />
buried epigenetic soils.<br />
The organic horizons often consist of shallow layers<br />
dry peat with high amounts of mineral material. The paleosol<br />
properties of the late Carga pedocomplex show a gradual<br />
trend towards harsher climatic conditions in this period.<br />
Synlithogenetic pedogenesis<br />
The cryopedolite forms under the influence of synlithogenetic<br />
soil formation. There is no differentiation of the activelayer<br />
material into genetic horizons. Concentrations of organic<br />
material are low and other soil properties weakly developed.<br />
Modern investigations of ice-loess deposits are conducted<br />
using spore-pollen analysis along with paleontologic and paleopedologic<br />
studies. For reconstruction of synlithogenetic soil<br />
formation the use of these methods is not sufficient. During the<br />
past years, ground squirrel burrows of the subgenus<br />
Urocitellus and of other rodents (lemmings and mice) were<br />
found in Late Pleistocene ice-loess deposits of North-East<br />
Eurasia. These objects differ with respect to depth of location<br />
inside the deposits, to structure and to content of frozen mate<br />
rials. The radiocarbon age of burrows is about 32 to 28 thou-<br />
191<br />
d
Zanina, 0. G. et al.: Pedogenesis of carga therrnochxane paleosols ...<br />
sand years. The formation of the burrows thus relates to the of epigenetic soil formation were characterised by a decreas-<br />
Carga thermochrone. The analysis of contents of fossil bur- ing intensity of eolian sedimentation, a slow increase in hurows<br />
and the ice-loess deposits containing them enables an midity and a warm climate. Synlithogenetic soil fornation<br />
impression to be gained about the environments of corre- took place under conditions of dry tundra landscape with<br />
sponding local habitats.<br />
steppe vegetation as well as active accumulation of eolian<br />
The structure of a number of fossil ground squirrel burrows deposits at the surface.<br />
has allowed to determine the thickness of the active layer at<br />
60 to 80 cm during the time of their construction. The presence<br />
of sublimation ice in the central parts of chambers ACKNOWLEDGEMENTS<br />
proves the absence of their flooding and low moisture of the<br />
bottom parts of the active layer during maximum thaw. Uni- We thank Dr. S.V. Kiselev and Dr. G.G. Boeskorov for help<br />
form distribution of burrows within Late Pleistocene polygons with the identification of material collected from fossil burrows.<br />
allows to assume plairl character of a surface near the bur- This original research was supported by the Russian Founrows.<br />
The simple structure of burrows, especially the ab dation for Basic <strong>Research</strong>, grants 01-05-64496,01-04-49087.<br />
sene of branching and platforms of excrements close to the<br />
entrance, points to a short-term existence of these constnrctions<br />
and rapid blocking of the surface by deposits from active REFERENCES<br />
eolian processes.<br />
Fossil burrows of ground squirrel contain organic matter in Gubin, S.V., Maksimovich, S.V. and Zanina, O.G. <strong>20</strong>01. Composition<br />
excellent condition and obviously introduced from the surface. of seeds from fossil gofer burrows in the ice-loess deposits of<br />
Zelyony mys as environmental indicator, Russia, Earth<br />
Remains of higher plants, including seeds and fruits, chitin<br />
Cryosphere, 5 (2): 76-82.<br />
remains of insects and their larvae, eggshells, feathers, ex- Gubin S.V. 1994. The Late Pleistocene soil formation at the coastal<br />
crements, bones and woolly mammals as well as their tis- lowlands of northern Yakutia, Russia. Soil Science 8: 5-14.<br />
sues were found among the contents of fossil burrows. The<br />
wool of large mammals found in burrows belongs to bisons<br />
and musk oxen which are widely distributed in tundra-steppe<br />
landscapes.<br />
The amount of seeds and fruits from higher plants can reach<br />
hundreds of thousands of units inside feeding chambers. The<br />
great volume of fruit and seeds from fossil burrows belongs t3<br />
the family of Brassicaceae, Ranunculaceae, Cariophy//aceae,<br />
Poaceae and Carex. TheseedsSilene sfenophylla Ledeb,<br />
Bisforta vivipara (L.) S.F. Gray, Potentilla nivea L and other<br />
species of sedges were found more frequently in burrows.<br />
The plant seeds from buried rodent burrows show the development<br />
of predominant tundra vegetation during the Carga<br />
thermocrone with the existence of pioneer and steppe species<br />
(Gubin et al. <strong>20</strong>01).<br />
Among the remains of fossil insects, various species<br />
such as leaf-beetle (Chrisolina rufilabris Fald, Ch. brunnicornis<br />
bermani Medv., Galeruca dahurica Jeann), billbug<br />
(Stephanocleonus eruditus Faust, S. fossilatus Fisch, Coniocleonus<br />
cf. ferrugineus Fahr) and larvae of several species d<br />
fly were discovered. These species are typical for the lower<br />
course of the Kolyma River during Late Pleistocene times.<br />
Almost all species are connected to steppe conditions. Collops<br />
obscuricornis Motsch (beetle of the family of Melyridae)<br />
was found in the west Chukotka area with modem relict<br />
steppe. This confirms that dry steppe conditions existed the in<br />
Northeast of Russia during Late Pleistocene time periods.<br />
CONCLUSION<br />
The Carga thermochrone in the investigated area is characterized<br />
by changes from synlithogenetic to epigenetic soil<br />
formations. Epigenetic soils buried during the Carga theme<br />
chrone from 40 to 28 thousand years ago provide information<br />
for reconstructing conditions of soil development. The periods<br />
192
8th International Conference on Permafrost Extended Abstracts on Current <strong>Research</strong> and Newly Available Information<br />
Changes in permafrost distribution and active layer th 'ckness in Canada<br />
during the <strong>20</strong>th century: a model assessment of clima :e change impact<br />
Y. Zhang, W. Chen, J. Cihlar, *S.L. Smith and *D. W. Riseborough<br />
Canada Center for Remote Sensing, Nafural Resources Canada, Offawa, Canada<br />
*Geological Survey of Canada, Natural Resources Canada, Ottawa, Canada<br />
INTRODUCTION<br />
Paleoclimate records and instrumental measurements show<br />
that the Arctic during the <strong>20</strong>th century is the warmest in the<br />
past four centuries, and air temperature in the northern high<br />
latitudes has increased at a higher rate than the global mean.<br />
The Arctic is extremely vulnerable to climatechange,and<br />
potential impacts include changes in permafrost distribution<br />
and active layer thickness, hydrological dynamics and carbon<br />
sinklsource, which may positively feedback to the climate<br />
system. This paper presents our recent results about the<br />
transient changes in permafrost distribution and active layer<br />
thickness in Canada during the <strong>20</strong>th century simulated by a<br />
process-based model at 0.5-degree latitudellongitude resolution.<br />
scale down this monthly dataset to half-hourly intervals<br />
(Chen et al. submitted). Other input data include land cover<br />
types, leaf area index, surface and soil organic matter, soil<br />
texture, the geothermal flux and ground ice content. They are<br />
converted to 0.5-degree latitudellongitude, corresponding b<br />
the spatial resolution of climate data. The initial conditions are<br />
determined by iteratively running the model for an initial year<br />
until annual soil temperature is stabilized.<br />
METHODOLOGY<br />
A process-based model to simulate permafrost thermal re<br />
gimes was developed by combining the strength of existing<br />
permafrost and land surface process models. Soil temperature<br />
and active layer thickness (ALT) were determined by<br />
solving the heat conduction equation, with the upper boundary<br />
conditions being determined using the surface energy balance,<br />
and the lower boundary conditions being defined as the<br />
geothermal heat flux at a depth of 35 m, where heat flux is not<br />
very sensitive to climate variations. We assumed that the<br />
heat flux at this depth does not change with time during the Figure 1. The components of the system considered in the model,<br />
simulation period, The model integrated the effects of climate, and energy fluxes between soil, vegetation and the atmosphere.<br />
vegetation, soil features and hydrological conditions based on<br />
energy and water transfer in the soil-vegetation-atmosphere<br />
system (Fig.1). The model was validated against the meas- RESULTS AND ANALYSIS<br />
urements at four sites in Canada. The simulation results were<br />
in agreement with the measurements of energy fluxes, snow Figure 2 shows the simulated spatial distribution of ALT and<br />
depth, soil temperature and thaw depth (Zhang et al. submit- its change. It also shows the distribution of permafrost areas<br />
w .<br />
and their change as well. The simulated permafrost distribu-<br />
Using this model, we simulated the transient response d tion is similar to the permafrost map of Canada (Heginbottom<br />
permafrost distribution and ALT in Canada to climate change et al. 1995), which is delineated based on permafrost site<br />
in the <strong>20</strong>th century. The climate data were obtained from a measurements, physiographic information and air tempera-<br />
0.5degree latitudellongitude gridded monthly dataset from ture. (It largely represents permafrost conditions in the recent<br />
1901 to 1995 (New et al. <strong>20</strong>00). A scheme was developed b decades, but we show it in all the panels for spatial refer-<br />
193
~~~~<br />
~.~<br />
some expansions in~iermafrost in these areas. These spatial<br />
distribution patterns correspond to the change in patterns of air<br />
temperature. Average ALT increased by 0.3 m for grid cells<br />
where ALT increased, and decreased by 0.1 m for grid cells<br />
where ALT decreased (excluding grid cells where pernabst<br />
disappeared or formed over this period). For all the permafrost<br />
grid cells ALT increased an average of 29% from the 1900s<br />
to the decade 1986-1 of 995.<br />
CONCLUSIONS<br />
Using a process-based model and inputs from climate re<br />
cords, soil features and remotely-sensed vegetation parameters,<br />
the transient response of permafrost distribution and aG<br />
tive layer thickness in Canada to climate change during the<br />
<strong>20</strong>th century was simulated. The results show that the land<br />
area underlain by permafrost decreased by 3% and ALT increased<br />
by 29% during the <strong>20</strong>th century. However, permafrost<br />
degradation and changes in ALT differ with space and<br />
time.<br />
REFERENCES<br />
Chen, W., Zhang, Y., Cihlar, J., Smith, S.L. and Riseborough, D.W.<br />
submitted. Changes in soil temperature and active layer thickness<br />
during the <strong>20</strong>th century in a permafrost region in western<br />
Canada. Journal of Geophysical <strong>Research</strong>.<br />
Heginbottom, J.A., Dubreuil, M.A. and Harker, P.A. 1995. Canada<br />
Permafrost. In; National Atlas of Canada 5th edition, Natural<br />
Resources Canada, Ottawa, Canada.<br />
New, M.G. Hulme, M. and Jones, P.D. <strong>20</strong>00. Representing twentieth<br />
century space-time climate variability, Part II: Development of<br />
1901-1996 monthly terrestrial climate fields. Journal of Climate<br />
13: 2217-2238.<br />
Zhang, Y., Chen, W. and Cihlar, J. submitted. A process-based<br />
model for quantifying the impact of climate change on permafrost<br />
thermal regimes. Journal of Geophysical <strong>Research</strong>.<br />
194
Author Index<br />
Abramov A. A., 1, 1 15<br />
Ahonen L., 141<br />
Akerman J. H., I 9<br />
Allard M., 37<br />
Ammann W., 95<br />
Arenson L. U., 71<br />
Ariuntsetseg L., 45<br />
Asano K., 13<br />
Auerswald K., 135<br />
Avian M., 3<br />
Balobajev V., 5<br />
Baron L., 89<br />
Baroni C., 145<br />
Barry R. G., 21, 123<br />
Batbold A., 163<br />
Beer C., 7<br />
Beylich A. A,, 9<br />
Blomqvist R., 141<br />
Borzotta E., I I<br />
Brabham P., 139<br />
Brouchkov A., 13<br />
Brown J., 169, 153, 113, 123<br />
Buldovich S. N., 1<br />
Burgess M. M., 153<br />
Bykhovets S., 21<br />
Cai M. F., 97<br />
Campbell S., 139<br />
Carton A,, 145<br />
Celi L., 39<br />
Chekhina I., 15<br />
Chelli A., 17<br />
Chen W., 193<br />
Chestnykh O., 103<br />
Christiansen H. H., I 9<br />
Chudinova S. M., 21<br />
Cihlar J., 193<br />
Dankers R., 85<br />
Davydov S. P., 41<br />
Degnan P., 141<br />
Delaloye R., 23, 93, 107, 89, 173<br />
Domnikova E. A., <strong>25</strong><br />
Drozdov D. S., 27,149<br />
Ednie M., 29<br />
Eiken T., 181<br />
Elliott C., 31<br />
Enkhtuul M., 163<br />
Etzelmuller B., 45<br />
Fabre D., 33<br />
Federici P. R., 33<br />
Fishback L-A., 35<br />
Fortier D., 37<br />
Frappe S. K., 141<br />
Frauenfelder R., 155,45<br />
Freppaz M., 39<br />
Frolov A. D., 67<br />
Fukuda M., 53,13<br />
41,75<br />
Gakhova E. N., 187<br />
Gilichinsky D. A., 21,75, 115, I, 41, 147<br />
Fyodorov-Davydov D. G.,<br />
Gintz D., 9<br />
Gorshkov S. P., 43<br />
Goulden C., 45<br />
Gruber S., 47,83, 77,65,59<br />
Gubin S. V., 75,41, 147, 191, 187<br />
Gubler H., 61<br />
Gude M., 7,111<br />
Haeberli W., 49,95, 1<strong>25</strong>, 173<br />
Hallet B., 151<br />
Hamano Y., 157<br />
Hanson S., 51<br />
Harada K., 53<br />
Harris C., 139<br />
Hauck C., 55,57<br />
Heggem E. S. F., 45<br />
Heiner S., 59<br />
Herz T., 61, 129,77,65<br />
Hinkel K. M., 113<br />
Hjort Jan, 63<br />
Hoelzle M., 47, 11 1, 51, 173<br />
Hof R., 65, 77<br />
Hollister R. D., 165<br />
Huggel C., 49<br />
lshikawa M., 67<br />
lstratov V. A,, 69<br />
Jambaljav Ya., 45, 163<br />
Jensen M., 141<br />
Johansen M. M., 71<br />
Kaab A., 155,45,183,1<strong>25</strong>,49<br />
Kadota T., 67<br />
Karelin D., 189<br />
Kasymskaya M., 73<br />
Kholodov A. L., 75, 1,41
King L., 77, 129, 61, 65<br />
King R. H., 35<br />
Kizya kov A. I., 79<br />
Kneisel C., 81, 55<br />
Kobabe S., 175<br />
Kohl T., 83<br />
Kolstrup E., 9<br />
Korostelev Y. V., 27<br />
Koster E. A., 85<br />
Kotov A,, 189<br />
Kristensen L., 87<br />
Krummenacher B., 173<br />
Kuchmin A. O., 67<br />
Kudryavtsev S., 167<br />
Lambiel C., 89, 93<br />
Laurinavichius K. S., 1<br />
Lehto K., 141<br />
Leibman M. O., 105<br />
Lewkowicz A. G., 29<br />
Linde N., 9<br />
Lisyuk M., 167<br />
Little J., 91<br />
Lucht W., 7<br />
Lugon R., 93,107<br />
Luetscher M., 143<br />
Lutschg M., 95<br />
Ma Q., 97<br />
Ma W., 97<br />
Makarov V. N., 99<br />
Malkova G., 103<br />
I01<br />
Marchenko S. S.,<br />
Mazhitova G., 103<br />
Meier E., 59<br />
Melnikov E. S., 105<br />
Metrailler S., 107<br />
Mergelov N. S., 41<br />
Mitusov A. V., I09<br />
Mitusova 0. E., 109<br />
Molau U., 9<br />
Monbaron M., 23,143<br />
Monnet R., 89<br />
Moren L., 141<br />
Moskalenko N. G.,<br />
105,169,177<br />
Mustafa O., 111<br />
Nelson F. E., 113,91<br />
Nikiforov S., 127<br />
Novototskaya-Vlasova K., 115<br />
Odegird R. S., 181<br />
Oelke C., 117<br />
Ogorodov S., 119<br />
Ohata T., 67<br />
Osokin N., 49<br />
Ostroumov V. E., 41<br />
Paananen M., 141<br />
Panova Y., 121<br />
Pappalardo M., 33, 17<br />
Paramonov V., 167<br />
Parsons M. A., 123<br />
Paul F., 1<strong>25</strong><br />
Pavlidis Y., 127<br />
Pedersen L. B., 9<br />
Perednya D. D., 79<br />
Perekalin S. O., 67<br />
Pfeiffer E."., 175<br />
Philippi S., 129<br />
Phillips M., 39, 135<br />
Polkvoj A,, 49<br />
Popova Alexandra A,, 131<br />
Puigdomenech I., 141<br />
Rachold V., 127<br />
Raynolds M. K., 177<br />
Razzhivin V., 189<br />
Repelewska-Pekalowa J., 19<br />
Reynard E., 93<br />
Ribolini A., 133, 33, 17<br />
Rist A. , 135<br />
Riseborough D., 193<br />
Rivkin F. M., 15<br />
Rivkina E. M., 75, I<br />
Rodriguez J. A. P., 137<br />
Romanovskii N., 73<br />
Romanovsky V., 153,45<br />
Ross N., 139<br />
Ruskeeniemi T., 141<br />
Sandall H., 91<br />
Saruul N., 45<br />
Sasaki S., 137<br />
Seppala M., 63<br />
Seppi R., 145<br />
Sergueev D., 73<br />
Serrano E., 93<br />
Schlapfer D., 47<br />
Schlatter F., 143<br />
Schmullius C., 7<br />
Schnellman M., 141<br />
Sharkuu N., 67,45<br />
Shashkin K., 167<br />
Shatilovich A., 147<br />
Shcherbakova V. A., 1<br />
Shepelev V. V., 5<br />
Shiklomanov N. I., 113<br />
Shmakova L. A., 147<br />
149<br />
Sletten R. S., 151<br />
Skvortsov A. G.,
Smith S. L., 153, 193<br />
Soina V., I 15<br />
Sollid J. L., 181<br />
Sorokovikov V. A., 21 , 7541<br />
Stockli V., 39<br />
Stoeckli V., 95<br />
Stoffel M., 143<br />
Streletskaya I. D., 159<br />
Strozzi T., 155<br />
Sueyoshi T., 157<br />
Tanaka M., 13<br />
Tarasov A., 159<br />
Thyrsted T., 9<br />
Tishkova M., 43<br />
Tomita F., 13<br />
Trombotto D. T., 161, 11<br />
Tumentsetseg S., 45<br />
Tumurbaatar D., 163<br />
Tweedie C. E., 165<br />
Ukraintseva N. G., 159<br />
Ulitsky V., 167<br />
Undarmaa D., 163<br />
van der Linden S., 85<br />
Vasiliev A. , 169, 105<br />
Venevsky S., 171<br />
Vonder Muhll D., 173<br />
Wada K., 53<br />
Wagner D., 175<br />
Wagner U., 57<br />
Walegur M., 91<br />
Walker D. A., 177<br />
Walker H. J., 179<br />
Wangensteen B., 181<br />
Warren I. M. T., 185<br />
Webber P. J., 165<br />
Weber M., 183<br />
Wegmiiller U., 155<br />
Wikstrom L., 141<br />
Williams P. J., 185<br />
Yashina S, G., 187<br />
Zamolodchikov D., 189, 103<br />
Zanina 0. G., 191<br />
Zhang T., 21 , 117,123<br />
Zhang Y., 193,67<br />
Zhang Z. H., 97<br />
Zimov S. A., 41<br />
Zotikov I., 49